I
L L
. . ' L . L > - ., ‘1‘ Tia.» Kit-$331.; .
.‘ L . _ ‘ ‘ g L L .- ~ L w s».: a.)
- , ' - ‘ , ‘ - " ~ .1»;ch ,
~ 7 ., . ~ . ~ ‘3: r, ,- .L Lag _ AC1 15,;
t ‘ " . - ‘ V ‘ : - ‘ ' ’ “.L‘ ' - . - , f #15: ‘..!‘.§:3.:,\- . uA
' ' ~ ' .. , . , t vu-
' ‘ ’ ‘ ‘ ' ‘ ‘ ‘ . . . w ,3. 1“-
. ' ~ 1 v
I
.
I
v
whim",
.
‘ «a
" (
v .
“a";
' ‘ ' L ' ‘ . I «my;
. ' u , . . ‘ ‘ ‘ . . . A 1 .1 ‘ ‘ “kit:
L V . , H.
. , _ X
a . ‘ '
,

.
f

rr' 2 "Ju‘r,

0 “’5'? “t

~ ~33.- 1.
. v.

1 .
. L175..- . v‘
H q . ,‘yyj “ m
t: L {find :, .
M’: n'tr; 1“,; .L
“\ ‘3‘: ‘. .

‘l

1'?

 

,. .
,. , . .- . M
kw}; 1.112;“;
”gs-2w W
:5. ma

a. M.

$5.41; 7
W333.

”VJ-”C" _
f ‘ 2",! @354

 

 

”a

UNIV

mmfliiijflijimuuiii'miiiiiu ”

‘KMH38447

This is to certify that the

dissertation entitled

Understanding Motion Picture Attendance
In The Age Of Home Video:
A Synthesis 0f Communication And Economic Approaches

presented by

Indrawansa de Silva

has been accepted towards fulfillment

 

of the requirements for

Ph.D. degree in Mass Nadia

 

\
r professor

Date 7’/§’q ‘f

MS U is an Affirmative Action/Equal Opportunity Institution 042771

 

 

LIBRARY
Michigan State
University

 

 

 

PLAcE It RETURN BOXtoromovothi-chockoummmm

TOAVOIDFINESrItunonorbdonddoduo.

DATE DUE DATE DUE DATE DUE

 

_.__~“_.._-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU loAn Animativo WW Opportunity Imtituion
Wanna-m

.~-———__~_A___

UNDERSTANDING MOTION PICTURE ATTENDANCE
IN THE AGE OF HOME VIDEO:
A SYNTHESIS OF COMMUNICATION AND ECONOMIC APPROACHES

By

Indrawansa de Silva

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Mass Media Ph.D. Program
College of Communication Arts & Sciences

1994

ABSTRACT

UNDERSTANDING MOTION PICTURE ATTENDANCE
IN THE AGE OF HOME VIDEO:
A SYNTHESIS OF COMMUNICATION AND ECONOMIC APPROACHES

By

Indrawansa de Silva

The purpose of this research is to explain motion picture attendance in
the context of a changing home video environment. The theoretical framework
for the current study is derived from the communication and economic
approaches that attempt to explain movie attendance. The communication
theory approach focuses on the individual decision making process of
moviegoing and traces the consumer decision making process through
expectancy-value theory, uses and gratifications. and diffusion of innovation

concepts.

The economic approach focuses on collective movie attendance decisions
and studies the effects of institutional factors in explaining movie attendance.
Both approaches share a similar set of predictor variables but use different
data sources and test different hypotheses. This study made an attempt to
converge these two approaches to derive a hybrid model to explain motion

picture attendance using household survey data.

Several multiple regression equations were estimated using data
obtained from a random sample of 366 Lansing area residents reached through
a telephone survey. The results found seven variables related to creative and

promotional aspects of the movies and two demographic factors as significant

predictors of movie attendance. The statistical flt (R2) of the equations ranged
from .1483 to .2069. explaining approximately 15 to 21 percent of the

variability in theater attendance.

Results of discriminant analysis revealed five creative and promotional
variables together with four demographic and economic variables as significant
discriminatory function variables for predicting the “frequent” and “infrequent”
moviegoers. These variables together explain about 21 percent of the
variability in the dependent measure (Lambda=.7891). The discriminatory
model correctly classified 69.45 percent of the cases. On the other hand. only
four demographic and economic variables were found to be significant
predictors of video rental patterns. The four variable model was able to explain

30 percent of the variability in video rental patterns (R2=.3003).

Overall. the results show some consistency with the studies that use
different data sources to explain movie attendance. Home video appears to
pose no immediate threat to the traditional theater but is likely to have an
impact on theater audiences as the clearance time between theater and video
windows continues to narrow. The movie industry. however. will be the
winner as the new technologies seem to help rather than hinder the economic

opportunities.

ACKNOWLEDGMENTS

Many fine individuals helped out my way through the completion of
doctoral work at Michigan State. First. I gratefully acknowledge the members
of my dissertation committee Dr. Barry Litman. Dr. Tom Baldwin. Dr. Bella
Mody and Dr. Bonnie Reece for their contribution to this dissertation. All of
them worked toward one goal: to help me to produce the best possible piece of
research. I hope I lived up to their expectations. My special gratitude goes to
Dr. Barry Litman for taking a keen interest in my work and for showing me the
joy of studying media economics. He is a dedicated teacher and a fine human

being most of us are not fortunate to meet often in our lives.

I would also like to thank Dr. Bruce Vanden Bergh. Chairperson of the
Department of Advertising. for going out of his way to help me on many
occasions. My life as a doctoral student would have been miserable without
Dr. Vanden Bergh‘s selfless help. Dr. Keith Adler was always there as a
teacher. colleague and a friend. Keith. I thank you very much! Nancy. Gloria
and Connie. your welcoming smiles and eagerness to help made me want to go

to the main office one more time.

It is also time to thank several of my friends and colleagues who made
all those long years less stressful. Larry Collette. your sense of humor and
being there to listen whenever I pick up the phone made me forget the fact

that I am thousands of miles away from home. Jaye. Leela. Marilyn and Rob

and Sandy. it would have been very difficult to come this far without good

friends like you.

Above all. a special thanks to my wife Priyani and my two daughters.
Malika and Samantha. for their unconditional love and patience which made
this whole thing possible. It seems futile to find words to describe how much i

treasure your love.

Table of Contents

LIST OF TABLES
LIST OF FIGURES ix
CHAPTER 1 Introduction 1
Current Status of the Motion Picture Industry ............................ 5
Movies As An Entertainment Option - Past and the Present .......... 6
The Postwar Period ................................................................ 12
Television and Motion Pictures ................................................. 18
Box Office: The Primary Window .............................................. 23
The Other Windows ............................................................... 25
Videocassettes ...................................................................... 26
Pay-Per-View ........................................................................ 32
Video-On-Demand ................................................................. 35
Laser Discs .......................................................................... 37
Foreign Markets .................................................................... 39
Summary ............................................................................. 47
CHAPTER 2 Motion Picture Attendance Research: A Review 49
Introduction ......................................................................... 49
The Communication Research Approach to Movie Audience .......... 51
Why Go to the Movies? ........................................................... 52
Expectancy-Value Theory and Its Applications to Moviegoing ......... 53
Uses and Gratifications Approach to Movie Attendance ................ 54
Shortcomings in Expectancy-Value Theory and Uses and
Gratifications Approach to Movie Attendance ....................... 56
Why A Particular Movie? ......................................................... 61
The Economic/ Business Approach ........................................... 66
Towards A Hybrid Approach ................................................... 7 O

CHAPTER 3 Research Questions and Methods 75

Research Questions ............................................................... 75
Selection of Variables ............................................................. 77
Creative and Promotional Variables ........................................... 79
Demographic Media and Economic Variables .............................. 83
Sample and Data Collection ..................................................... 85
Telephone Survey Outcome ..................................................... 86
CHAPTER 4 Results 88
Sample Demographics .......................................................... 88
An Overview of the Movie Attendance Patterns ......................... 90
The Social Context of Movie Attendance .................................. 94
The Other Windows: Watching Movies At Home ....................... 98
Results of Multivariate Analyses ............................................. 101
Data Reduction Efforts ......................................................... 106
Regression Analysis Results for theater Attendance ................... 107
Discriminating the “Frequent” and “Infrequent” Moviegoer .......... 1 15
CHAPTER 5 Discussion and Conclusions 123
Theater Attendance ............................................................. 123
Determinants of Theater Attendance ....................................... 124
Distinguishing “Frequent” and “Infrequent” Moviegoers .............. 128
Video Rental ....................................................................... 130
Determinants of Video Rental ................................................ 133
Limitations and Suggestions for Future Research ...................... 136
APPENDICES 140
REFERENCES 149

Table

1.1

1.2
1.3
1.4
1.5
1.6
3.1
4.1
4.2

4.3
4.4
4.5
4.6
4.7
4.8
4.9

4.10
4.11
4.12
4.13
4.14

List of Tables

Theater Attendance and Share in the Recreational and Spectator
Amusement .......................................................................

Weekly Movie Attendance in Relation to Television Penetration 16

Domestic Movie Revenues and Shares ................................... 21
Unit Transaction and Growth Per VCR Household ................... 31
10 Top Grossing Movies - 1992 ............................................ 41
Foreign Revenues and Shares for Movies ............................... 43
Telephone Survey Outcome .................................................. 87
Sample Demographics ......................................................... 89
X? .5532: °.‘. 95???? and 5355???? 13???? 3?. MW“ ............ 93
Least Like About Going To A Theater to See A Movie ................ 95
Most Important Reason For Going to A Theater ....................... 97
Factors Attracted to the Last Movie/Video .............................. 98
Most Important Reason for Renting A Movie ............................ 101
Correlation Matrix .............................................................. 103
Factor Matrix ..................................................................... 147
2331:3333??? Xi??? 13.5??? if??? .i??.?3?ff§?‘fff§f’f¥ 109
Regression Equations For Movie Attendance ........................... 1 11
Tolerance and Variance Inflation Factors ................................ 148
Discriminant Analysis of Movie Attendance ............................. l 17
Classification Results: Discriminant Analysis ........................... l 18
Regression Equations for Video Rental .................................... 120

VIII

List of Figures

1.1 Consumer Homevideo Spending 1985-1992 ............................ 21
2.1 A Propositional Model of the Movie Selection Process ................. 65

4.1 All Groups Stacked Histogram: Canonical Discriminant Function . 1 18

0:1
ho:
0w

ex;

Chapter 1

Introduction

Recent changes in the home video environment have drastically altered
the way people see movies. Evidently. home video affects the decision of what
movie to see. where to see it. and when to see it: thus it has a profound effect
on the exhibition side of the motion picture industry. The rapid growth of
home video technologies now allow people to get theater-like experience in their
own living rooms at ever affordable prices. The home video market
experienced the fastest growth of any product or service in the
communications industry during the past few years. In fact. by 1989.
spending on home video comprised 40 percent of all US. spending on filmed
entertainment. while box-office revenues only accounted for 21 percent

(Veronis. 1992).

VCR penetration in the US. has nearly tripled over the last six years.
Currently over 80 percent of US. households are equipped with a VCR. and it
is estimated that the VCR penetration will reach 88 percent by 1995 (New York
Times. February 4. 1993. p. B4). If the industry forecasts materialize. VCR
penetration in the United States will approach total television penetration by

the end of the century (Veronis. 1992). To meet the growing demand for

50

TC]

47

SC]

5111

113'

50‘.
I65

Lid

{0 1'

W31:

2

software created by the changes in the infrastructure. the number of video
rental stores rose from a mere 2.500 in 1980 to a phenomenal 60.000 in
1990. During the latter half of the 1980s. home video spending soared at a
47.5 percent compound annual growth rate. the fastest of any product or

service in the communications industry in the U.S. (Veronis. 1992).

According to industry statistics. rental of prerecorded movies currently
surpasses all available means of watching a movie (Quigley. 1992). A recent
nationwide study by the industry revealed that 67 percent of the people prefer
to watch movies at home rather than go to a theater (Nichols. 1992a). The
most recent industry statistics seem to confirm the survey results. For
example. in 1992 consumers rented more than four billion videotapes and
spent nearly $12 billion to rent and purchase prerecorded Videocassettes. In
response to the growth of technology and consequent demand for software. the
industry has promptly narrowed the passage of time between the theater and
video releases to an extent that some movies now practically step from theater
to video stores. By late 1991. the passage was reduced to a mere two months.

which is needed for distributors to promote titles and stores to place orders.

The financial weight of the video release has become so crucial that
some major studios are now sending movies directly from theaters to video
stores. For example. Fox decided to release the video of Mrs. Doubtflre on April
26. 1994. just about five months after its theater release. What distinguished
Fox‘s action from the convention is that at the time of the video release Mrs.
Douby‘ire ranked number 10 on Variety's domestic box-office list. drawing
nearly $2 million per week in receipts from 1.153 theaters. While the video

release effectively ends the theater run. senior vice president for Fox. Bruce

3

Pfander. defended the decision by saying. “We may lose a few million on the
theatrical side. but with video we're pumping in another $100 million” (New
York Times. April 8. 1994. p. B4). Justifying such marketing strategies. a
video industry expert noted: "the shorter the time is good because the film is
fresh in people's mind. and with it just coming out of theaters you get more

bang out of the theatrical advertising" (Nichols. 1992b).

As a result. the near century domination of the traditional theater as the
primary window for movies has been effectively threatened by the thriving
home video industry. Financially theatrical revenues now lag way behind the
video revenues. here in the U.S. and in the global market. Apart from the
social/ psychological gratifications one could gain by going out to see a movie.
the benefits of watching movies at home appear to far outweigh the theater-

going experience. As George Lucas succinctly observed:

For years I've had arguments with executives. and I've been saying

things are going to change rather dramatically and it has to do with the

cassette industry. Once you could get a movie for two bucks six months
after it was released to theaters. it had to basically change the business.

Why would somebody want to go to movies now and spend 20 or 30

bucks with their family. when they can spend 2 bucks in a couple of

months? It's had a profound effect on all of us (New York Times.

January 27. 1992 p. Bl).

Unlike the early days. where theatrical success was positively related to
television performances. the video market seems to be somewhat independent
from the theatrical performances. Many recent releases. even with name
stars. that did not do well in the theaters. seem to perform well in the video
stores (eg. Career Opportunities. King Ralph. The Field. La Femme Nikkita. True

Colors. to name a few). This suggests that. as George Lucas observed. a

significant segment of the audience is now willing to wait for movies to come on

 

tide!

Sr-‘V‘

he:

num

4
video to see them for the first time. This may be a partial explanation for the

sagging theater attendance that has fluctuated around one billion a year despite
the substantial population grth and the reported 35 percent increase in the

number of movie theaters.

Although the local theater still remains the “initial” window for feature
length movies. its share in the total gross appears to have changed significantly
and needs to be treated accordingly when analyzing the motion picture
audience and its economics. It is timely to ask then: How do these structural
changes in the home video landscape influence our efforts to explain motion
picture attendance? The purpose of this dissertation research. therefore, is to
explain motion picture attendance in the context of rapidly changing home
video environment. Specifically. this research will attempt to answer the

following two research questions:

Research Question 1: What factors have the most influence when
consumers are deciding which movie to see?

Research Question 2: What are the efl‘ects of the growing home video
environment on movies seen in general and theater attendance in

particular?

It seems premature at this point to justify the study and provide a
rationale for these research questions without discussing the current status of
the movie industry and reviewing appropriate past research that will provide
theoretical guidance for this research. As such. the following section will
discuss the current status of the movie industry. followed by a chapter that
reviews literature pertinent to the two theoretical frameworks used in this

study. A detailed discussion of the research questions. the selection of the

variables. sample and the methods will be presented in chapter 3.

5
Current Status Of The Motion Picture Industry

Americans spent 4.9 billion dollars at the box-office buying 971.2 million
tickets in 1992‘. And 1992 was the 31st consecutive year in which U.S. film
ticket sales varied around the one billion markz. In today's multi-window
market. where movies are being delivered to the public through a vast array of
traditional and non-traditional channels. box-office grosses and ticket sales
alone hardly provide a complete picture of the movie industry and its
economics. Movies are a part of the 25 billion dollar entertainment industry
where the theater has become just one of the many exhibition windows through
which movies are delivered. Although motion pictures remain a healthy
industry. its share of the total spectator amusement expenditure has dropped
from an impressive 77 percent to a mere 4 percent over the past 40 year
period3 (Quigley. 1991). Once limited to the theater and network television.
movies are now the staple of the multi-billion dollar home video industry that
includes videocassette. cable TV. pay-cable. pay-per-view. video-on-demand
and laser disc. As such. the economic and social significance of the theater
has drastically changed over the past decade and continues to do so as new

windows for movies flourish in an unprecedented rate. It is in this context that

 

1 A two percent increase in box-office gross over 1991's 84.8 billion but a 1.1% decrease in
number of tickets sold (981.9 million tickets were sold in 1991).

2 Lowest year since 1961 was 1971 with 820.3 million tickets sold and the highest year was
1984 with 1.2 billion tickets sold. The 1993 figures. however. were predicted to be record
setting due to the several very successful summer releases and factors related to the video
industry (Variety. November 29. 1993. p. 1)

3 Share of the motion picture industry in the total spectator amusement expenditure 1950-87:
1950 - 77.26%

1960 - 59.22%
1970 - 48.00%
1980 - 4.27%
1987 - 3.69%

ti

'1:
u“

6

this chapter attempts to review the necessary history and the development of

the movie industry to its current status.

Movies As An Entertainment Option - Past And The Present

It was a century ago. on April 14. 1894. that the world's first
Kinetoscope parlor opened its doors in New York City. This historic event
marked the beginning of motion picture industry in the U.S. and had
drastically altered the way people spend their leisure time in the twentieth
century. The Klnetoscope became such an instant success that within a few
months Kinetoscope machines were installed in department stores. drug
stores. hotels. bar-rooms and phonographic parlors in almost all major cities

throughout the country (Hendricks. 1985).

The popularity of Kinetoscope. however. was short lived for two major
reasons. first. the initial attraction to Kinetoscope was due mostly to its
technological novelty. which had a short life span. rather than its entertainment
value. Second. Kinetoscope was only able to entertain one viewer at a time.
consequently making it economically not very attractive. The one-machine-per-
person hardly allowed Kinetoscope to stand as a self-supporting medium.
Except for the few Kinetoscope parlors in some big cities. they were placed in
other commercial or entertainment establishments. As such. once the novelty
wore off. Kinetoscope failed to attract a commercially supportive audience
necessary for its economical survival. But the initial success of the
Kinetoscope indicated the commercial potential of moving pictures. As Jowett
(1976) noted. the Kinetoscope acted as “an incentive to enterprising inventors
and businessmen to devise a machine that would project the moving pictures

on a screen where they could be viewed by a larger number of people at one

7
time.” Driven by the promise of large financial rewards attached to large

audiences. In less than two years the Kinetoscope was promptly pushed aside
by a technologically superior new innovation called the Vitascope - embryo of

the modern day movie projector (Hendricks. 1985).

While Edison concentrated on the camera. others. especially the
Lumiere brothers of France. envisioned the projector as an important part of
the entire process of moving pictures (Gomery. 1991). As a result of the
collective technological efforts that took place in the U.S. and Europe. the new
machine-Vitascope-allowed picture images be projected onto a screen so that a
group of people. not an individual. could be entertained at one time. In an
industry where profits have always been the first and the foremost concern of
the companies involved. Vitascope provided the much needed economic
promise and assurance (Balio. 1985). Compared with Kinetoscope. Vitascope
was indeed a commercial and aesthetic breakthrough. It allowed
entrepreneurs to collect admission from a group of people from a single
screening. This. in fact. changed the course of the motion picture industry
economically and socially. As Guback (1987) noted:

The amount of revenue earned by a peephole parlor entrepreneur

depended upon how many machines could be in operation at once. It

was in this context that the introduction of the projector was so
earthshaking. Aesthetically. of course. it changed the way people
watched films. But its most significant impact. from the business
perspective. was that it vastly increased the rate at which investments
in films and equipment could be recouped.

The popularity of Vitascope brought the movie industry to its first “boom
stage.” Despite some difficulties experienced in production and distribution.
movies were established as a staple in vaudeville shows (J owett. 1976). The

industry also realized that its long term survival was largely dependent upon

8

the constant supply of new films that could attract the audiences to the
theaters on a regular basis. As a result. barely ten years after the introduction
of the peephole machine, nearly 150 film exchanges were in operation serving
all areas of the country. Balio (1985) called this the industry's conversion to
mass production of motion pictures. which. in effect. helped open thousands of
theaters all over the country. The exhibition boom gave birth to nickelodeon
theaters which. according to Merritt (1985). established a durable pattern for
nation-wide distribution. and-most importantly-built for the motion picture an

audience that would continue to support it for another thirty years.

The proliferation of the nickelodeon movie theater began to replace
single-reel travelogues and news films with narrative ones (Allen. 1985). And
movies had gone from simply recording events to story telling. They also
became inexpensive. easily accessible. and. before the introduction of sound.
did not require an understanding of spoken English to enjoy them - an
important factor given the large portion of immigrants in the audience. Movies
quickly surpassed the other dominant forms of entertainment. such as plays
and musicals. and became the most popular means of entertainment. More
people went to the movies than participated in any other form of commercial
recreation (Austin. 1989). By 1910. there were more than ten thousand
nickelodeon theaters in operation throughout the country “creating demands for
between one hundred and two hundred reels of film every week” (Merritt.
1985). Based on the most reliable estimates at the time. Merritt (1985) found
movie attendance practically doubled during the nickelodeon era. increasing
from twenty-six million persons per week in 1908 to at least forty-nine million

in 1914. It was this dramatic growth of nickelodeon. noted Jowett (1976). that

9

led to the development of the motion picture industry which consisted of three

basic segments-production. distribution and exhibition.

Large audience size notwithstanding. movies did not yet become a part
of the American entertainment and cultural mainstream. The public perceived
movies as a medium directed towards and appealing to a specific social class.
namely the urban working-class. of which the immigrants were a significant
segment. The public perception that movie entertainment was “lower class”
oriented shrunk the movie industry’s boundaries and its economic prospects.
In an attempt to change the public's image and the social stigma attached to
movies. the industry began to improve the movies and the theater quality as a
lure to attract other segments of the public to the theaters. And it did work.
As the film quality improved and theaters became more comfortable. noted
Merritt (1985). movies began to appeal to the larger middle and the upper
class. The audience was no longer made up primarily of the “urban working
class” and the “immigrant.” The movies that evolved from arcades to vaudeville
theaters and then to store-front theaters were soon shown in motion picture
palaces. They were able to attract the much needed middle-class audience.
Movies finally became a part of the mainstream American culture (Jowett.
1976). The growth of the audience was so rapid and dramatic. the drama
critic for the New York Times. reporting the occasion of opening of Strand
Picture Palace on April 1. 1914. wrote:

When I saw the wonderful audience last night in all its costly tags. the

one thought that came to my mind was that if anyone had told me two

years ago that the time would come when the finest looking people in
town would be going to the biggest and newest theater on Broadway for
the purpose of seeing motion pictures. I would have sent them down to

visit my friend. Dr. Minas Gregory at Bellevue Hospital. The doctor
runs the city’s bughouse you know. (Jowett. 1976. p. 60)

10

Once the members of all segments of the society were lured into the
theaters. the industry faced the challenge of keeping their enthusiasm to the
new medium intact. One of the most important steps taken by the movie
industry in this regard was the development of the “star system.” Although
previously unknown and insignificant. actors rose into the stardom and were
paid exorbitant wages. Of all the devices use to generate business. observed
Balio (1985). the star system was by far the most effective. In its fully
developed form. the star system affected all the branches of the industry. For
some. acting suddenly became the best-paid profession on earth. While the
average American industrial worker earned less than fifty cents an hour.
million dollar guaranteed salaries became standard for stars (Platt. 1992). For
example. as early as 1916 Charlie Chaplain collected a $10,000 per week
stipend and $150,000 in bonus money for signing. The studios embraced the
star system as a mean of developing a faithful audience. And the large sums
paid to the stars apparently seems justifiable because they provided the much
needed assurance in this unpredictable trade. From the beginning. name stars
were able to secure a profitable audience more than any other single factor.
By 1922. millions of people flocked to the theaters on a weekly basis. and the
motion picture industry had become the largest and most widespread

commercial entertainment form the world had ever seen (Jowett. 1976).

The prosperity. however. did not last long. By mid 19203 the mostly B-
class run-of-the-mill movies were not able to justify the rising admission prices
and high expectations of maturing audiences. As audiences now represented a
cross section of the society. demand for good movies rose high. While the

studios were desperately looking for some novelty to overcome the

11
lukewarmness and the silent antagonism. another technological breakthrough.

sound. came to the rescue of the industry. The arrival of sound was a shot in
the arm that the industry so urgently needed in the late 1920s (J owett. 1976).
With the adaptation of sound. the remainder of nickelodeon theaters gave way
to “true” theaters and the movie industry was able to revive and go through the

Depression as one of the least affected industries in the nation.

By 1930 the motion picture industry had become a mature oligopoly that
smoothly turned out films with regularity (Balio. 1985: Jowett. 1976). Five
major companies (Paramount. Wentieth Century-Fox. Warner Bros.. RKO and
Loew‘s). known as the “Big Five.” dominated all three segments-production.
distribution and exhibition-of the industry. Universal. Columbia and United
Artists. known as the “Little Three.” operated in a symbiotic manner with the
Big Five (Balio. 1985). The Little Three owned no theaters but their access to
the first run screens owned by the Big Five made them major companies.
These studios practically produced all class-A features that played in the best
theaters and generated the most revenues. In distribution. the eight majors
collected about 95 percent of all film rentals paid to national distributors. And
on the exhibition side. the Big Five alone owned or controlled the operations of
126 out of 163 first-run theaters in the twenty-five largest cities of the country
which accounted for nearly 70 percent of the nation's box office receipts
(Huettig. 1985). The big studios kept the competition to a most profitable level
to all the companies involved by competing and cooperating when and where it
was necessary. In this “community of interest.” observed Balio (1985). “a hit
motion picture was profitable for all the integrated companies." The vertically

integrated structure of the major studios lasted until the Supreme Court held

12

them responsible for antitrust violations and ordered them to divest the theater
holdings. The Supreme Court ruling and postwar demographic and
psychographic changes. along with the rise of television. brought the movie
industry to a crucial point in its history. The new era. which was appropriately
called the “age of television.” brought new challenges as well as opportunities to
the nearly half-a-century old industry that so far relied upon one outlet: the

theater.

The Postwar Period

United States was the only country in the war to come out richer than
when it went in. As a result. the postwar U.S. economy witnessed an
unprecedented economic growth coupled with an increase in leisure time. The
steady growth of GNP inflated the public's disposable income and recreation
expenditure. ideally a fertile ground for an industry like movies. Yet. the
motion picture industry began to confront a period of economic difficulties.
After seeing its biggest year in 1946. in which the average weekly audience hit
an unprecedented 90 million and box office gross accounted for $1.7 billion.
the audience started to erode. During the fifteen year period between 1946
and 1960 the average weekly audience dropped from 90 million to 40 million
and the movie share of the total recreational expenditure dropped from one
fifth to one twentieth (see Tables 1.1 and 1.2). Movies continued as a major
entertainment medium in the American recreational mix but lost its place as

the major source of commercial recreation.

13

TABLE 1 .1
THEATER ADMISSIONS AND SHARE IN THE RECREATIONAL AND SPECTATOR AMUSEMENT

 

 

1935 - 1985

Share Of Total Share Of Total

Admissions Recreational Spectator Amusement

Year 6 Millions) Expenditures Expenditure
1935 556 21.14% ' 82.74%
1940 735 19.54 % 81 .31 %
1945 1,450 23.62% 84.60%
1950 1,376 12.34% 77.26%
1955 1,326 9.42% 73.63%
1960 951 5.20% 59.22%
1965 927 3.52% 51.19%
1970 1,162 2.86% 48.01%
1975 2.115 3.84% 51.81 %
1980 2,750 2.56% 42.7%
1985 3,749 2.43% 3.85%

 

SOURCES: Jowm 1976; QUIGLEY 1992

Although television evidently had an enormous impact on movie
attendance. the postwar erosion of the attendance started even before
television became a predominant entertainment medium in American society.
According to Jowett (1976). demographic. economic. legal. and lifestyle factors
combined with quality of the movies themselves and television contributed to
the postwar decline of movie audiences. As the post war population grew older
and more and more young people began to start families. movies began to lose
its largest group - the group less than 30 years old who comprised more than
50 percent of movie patronage (Conant. 1960). As Lazarsfeld (1947) pointed

out:

14

The decline of frequent movie attendance with increasing age is very
sharp. No other mass medium shows a comparable trend. This is
probably due to variety of factors. Movie-going is essentially a social
activity. and young people are more likely to band together for the
purpose of entertainment. Then. for the movies one has to leave home.
which probably becomes more distasteful as one grows older.
Meanwhile. the public's reaction to the movie quality appeared to grow
bitter while the entertainment options available to them grew bigger. For
example. a survey conducted by Fortune magazine in 1948 found 38 percent of
the respondents believing that there were “fewer good movies” in 1948 than
three years ago. A more disturbing finding of the Fortune survey. however.
was the indication that a majority (50%) of the respondents were beginning to
find alternatives to going to the movies (J owett. 1976). The popularity of other
alternative spectator amusement. such as professional sports. combined with
the acceptance of such outdoor activities as camping and boating also indicated
the defection from the movies. As is evident from the figures shown in the
Table 1.1. the theater admission share continues to decrease in both total

recreational expenditure and total spectator amusement expenditure.

The Justice Department meanwhile brought charges against the major
studios for defiance of the antitrust laws. The Paramount case I. filed in July
20. 1938. was finally settled by the Supreme Court in 1946 (Paramount case
11): it ordered the five major studios to divest their most profitable segment
(exhibition) from their production-distribution activities and enter into consent
decrees in other areas (Conant. 1960). The Paramount decrees came as a
major blow to the industry already in trouble and had an impact on its
structure. behavior. and performance. In the long run. however. noted Jowett.

“the divorcement decree was but one of the several social and economic factors

15

which together caused a total shift in the pattern of motion picture attendance

in the postwar period” (Jowett. 1976).

In addition to the ongoing problems in distribution and exhibition. the
production costs of movies continued to grow while profits from the only
ancillary market at the time-foreign sales-dwindled as a direct result of strict
measures taken by the European countries to restrict the flow of foreign
exchange. While the industry was struggling to get the audiences back to the
theaters. television. the newly emerging medium. provided the public with a
convenient alternative to moviegoing. With the proliferation of television. the
movie industry entered an era which saw a systematic and continuous decline
of the audiences. As Table 1.2 shows. by 1949 television became a true mass
medium in the U.S. During this year the number of television stations on the
air rose from 17 to 50 and the number of households with television receivers
exceeded the 1948 total by 500 percent while the movie attendance started to

decline.

Hollywood initially reacted to television merely as ”visual radio” with no
long term threat to movies. But it soon found that presumption painfully
incorrect. As seen in Table 1.2. there is a very strong inverse relationship
between television penetration and movie attendance“. Conant (1960).
described the impact of television on the allocation of leisure time as

“spectacular” and asserted that as a “conservative” description. After

 

4 A Pearson correlation coefficient calculated using this data shows a very strong inverse
relationship between the number of families with TV sets and average weekly movie attendance
(r=-.83 sigiificant at .001) and a stronger relationship between the number of TV households
and the number of TV stations (r=.95 significant at .001)

16

TABLE 1 .2

 

WEEKLY MOVIE ATTENDANCE IN RELATION TO TELEVISION PENETRATION

 

 

1946 - 1956
FAMILIES WITH TV NUMBER OF AVG WEEKLY MOVIE
SETS (’000) COMMERCIAL TV ATTENDANCE
Year STATIONS (MILLIONS)
1946 8 6 90
1947 14 7 90
1948 172 17 90
1949 940 50 70
1950 3875 97 60
1951 10320 107 54
1952 15300 108 51
1953 20400 125 46
1954 26000 349 49
1955 30700 411 46
1956 34900 442 47

 

SOURCES: Film: The Democratic Art, jowett, 1976

examining the introduction of television and movie attendance patterns. Stuart
(1975) also found empirical evidence to substantiate the hypothesis that the
introduction of television was the principal reason for the decline in movie
attendance. He then concluded: “In the absence of television competition.
there might have been a substantial increase in motion picture revenues
between 1948 and 1954.” The empirical evidence. in fact. only confirmed

what the movie industry had already been experiencing.

Within the first ten years of television. movies lost nearly half of its
audience evidently to the new medium (Table 1.2). Only a few in the film

industry seemed to realize the actual impact of television on the movies and

17

what could be done to rescue it from apparent disaster. One such visionary
was Samuel Goldwyn (1950). As he succinctly illustrated:
A factor on our side is that the people will always go out to be
entertained because human beings are naturally gregarious. But before
the movie-goer of the future arranges a baby sitter. hurries through
dinner. drives several miles. and has to find a place to park. just for the
pleasure of stepping up to the box office to buy a pair Of tickets. he will

want to be certain that what he pays for is worth that much more than
what he could be seeing at home without any inconvenience at all.

The implication of Goldwyn’s observation is that movies must be
substantially different from TV or they should offer something that TV cannot
offer in order to bring the audiences back to the theaters. Acting on this
premise. the industry evidently fought back first by offering color and then
moving to epics and technological gimmicks such as 3-D. CinemaScope.
Cinerama. stereophonic sound and Todd A-O. to distance movies from
television. Although technological wizardry was able to win some audiences
back. the industry was fighting an uphill battle against another more exciting
technological breakthrough. namely television. And the odds. however. were
against the movies with television offering “free” entertainment at the
convenience of one's own living room. Eventually the “all out competition”
failed. the movie industry started to cooperate and integrate with television
(Jowett. 1976). An economic symbiosis began between the two media.

Perhaps to its surprise the industry soon realized that there is money to be
made by feeding the voracious appetite of television with its large inventories of
old movies. Cooperation proved more lucrative than the competition. Once the
public signaled their willingness to stay home and enjoy the so-called “free”
medium. Hollywood started to flood television with its old movies. A special
report prepared by Sindlinger and Company for the Theater Owners of America

noted that in the last quarter of 1957. old movies constituted almost 25

18

percent of television viewing time (Jowett. 1976). It’s no surprise then that the
movie attendance dropped to an all time low of 37.7 million a week in the same
year (Conant. 1960). By 1958. an estimated 3.700 movies had been sold or

leased to television for an estimated $220 million (Jowett. 1976).

As Litman (1982) observed. in addition to selling theatrical movies. the
major studios became suppliers to the networks of all forms of programming.
ranging from prime-time and daytime series. As a result. by late 1950s
television became Hollywood’s biggest money maker. According to some
estimates a single production company. such as Desilu. turned out more
footage than the combined output of the five major studios (Jowett. 1976). The
economic symbiosis that began between the two media in the mid 1950s took
on a pattern of stability and mutual interdependence within a short span of
about fifteen years (Litman. 1982). It had also drastically changed the
economic structure of the movie industry by offering a strong ancillary market
until the 1980s when the revenues from other outlets began to surpass

television revenues.

Television and Motion Pictures

As seen in the preceding discussion. it was evident that television
contributed more than any other single factor to weakening the theater
audience. Ever since television permeated America in the 1950s. per capita
movie attendance dropped by about 75 percent (American Demographics.
September 1986. p. 60). Nevertheless. the new medium also proved to be a
strong secondary market for the movies. The poor quality of television
programming in the early days. combined with the television's great appetite for

programming. forced television executives to turn to Hollywood movies to fill

l 9
the air time. The truce between the two media started when Hollywood began

selling its old movies to television on mutually beneficial terms. In the
beginning of this economic symbiosis. any money the studios made by selling
TV rights to their mostly B-class movies was considered as windfall profits
because these libraries had been fully amortized. However, as the demand
increased. the studios quickly discovered that they had undervalued their old
films. As a result. during the early 19608, noted Balio (1985). a run-of-the-mill
feature leased to the networks fetched $150,000 for two showings. as
compared with an average price of $10,000 per film for the RKO library in

1955 (Balio. 1985). And the prices only went up.

According to Londoner (1985). during the 1960s the cost of a typical
movie to the networks for three-year license rose from $150.000 to $800.000.
fueled by the attempts of then young network ABC to gain some ground in the
rating game. Searching for a programming strategy to move up from the third
place in the network hierarchy. ABC decided to bet on theatrical movies and
started to bid for movies that had proved their mass appeal at the box-office
and was willing to pay above average prices for them. For example. when the
average movie was fetching $400,000. ABC paid Columbia $2 million for Bridge
on the River Kwai. And it paid off. When it was aired in September 1966. the
nine year old movie made TV history by attracting an unprecedented audience
of 60 million. It was able to beat such popular TV hits as the Ed Sullivan Show
and Bonanza in the ratings (Balio. 1985). Stirred by ABC's success. the other
two networks followed suit creating a bidding war for A-class movies. The
newfound competitive bidding war by the networks doubled the average price

of the movies from $400,000 to $800,000 by 1968 while theatrical classics.

20
such as Cleopatra and Gone With the Wind each fetched $5 million (J owett.

1976). As a result, by late 19603 television had become a firm secondary
market for theatrical movies. Except for a short period of time in the early
19703, when the prices dropped in half. movie rights for broadcasting
continued to grow. The prices gradually worked up to about $1 million by
1975. And by late 19703 movies were drawing close to $2.5 million in the

network market (Balio. 1985).

Although television revenues were first considered as just ”gravy" by the
studios, they soon became expected and planned for. As the television market
developed. studios started to use television revenues as collateral in obtaining
financing. Still. network premieres typically began about a year-and-a-half to
three years after the theatrical run (Balio. 1985). As Litman (1979) observed,
television networks had greater difficulty in negotiating with studios as
suppliers of theatrical films than as suppliers of other entertainment
programming due to the lesser uncertainty attached to the movies that had
already undergone the popularity test in the theatrical run. The growing power
of the studios forced the networks to start financing what became known as
made-for-television movies as a supplementary to the theatricals. This created
a period of time with seven movie nights on TV. The made-for-TV movies and
theatrical movies not only created a glut on the market but also made
audiences more selective and consequently dropped the ratings. It took
another four years - until 1972 - for the studios to tip the supply and demand
relationship vis-a-vis the networks in their favor. By this time theatrical
blockbusters and movies with bankable stars were able to draw lucrative

prices in the television market. For example. networks paid $7.5 million for

2 l
The Deep. $15 million for Allen and $10-$13 million for any Clint Eastwood or

Burt Reynolds film (Balio. 1985). This trend continued until the late 19703
when the new technological developments such as cable TV . Videocassettes,
pay-TV. superstations and independent stations emerged as new windows to
offer motion pictures. These new windows of the late 19703 that grew during
the 19803 opened more markets for movies and changed the economic
parameters of the movie industry. As seen in Table 1.3. the broadcast
television share of the domestic movie revenues dropped from 12 percent to 8
percent during the period of 1985 to 1990 and had the lowest annual growth

compared to box-office. pay-TV and videocassette for the same period.

 

 

 

Table 1.3
DOMESTIC MOVIE REVENUES AND SHARES

Compound
1985 1990 Annual
Category (5 Millions) Share (S Millions) Share Growth
Box Office Rentals 1.1612 42% 2,160 32% 6%
Television 1,074 28% 1,642 24% 8.9%
Broadcast TV 442 12% 569 8% 5.2%
Pay TV 632 16% 1,073 16% 11,2%
Videocassettes 1,136 30% 2,925 44% 20.8%
Total 3,822 100% 6,727 100% 12.0%

 

Source: Veronis, Suhler & Associates. Communication Industry Forecast, July, 1992

The large home video and pay-cable audiences to which movies are first
exposed sharply reduced the network audience for theatrical movies. As
ratings dwindled, the networks expressed their reluctance to pay as much as
in the past for feature film exhibition rights (Vogel. 1986). The networks also
began to rely less and less on theatrical movies in their program planning. The

programming trends of the recent past show the networks building audiences

22
around docudrama type made-for-TV movies with stories still afresh in the

audience’s memory and on popular pre-sold stories which are both less
expensive to buy/ produce than the theatrical movies but more effective in
generating higher ratings. The networks. however, still remain as an important
market for theatrical movies. But as figures in Table 1.3 indicate, its economic
importance to the movie industry is largely shrinking. Television now accounts

for only about one-fourth of the domestic movie revenues.

The evidence also suggests that the networks are becoming more
selective about the movies they put on the air. Unlike the days of blind
bidding where movie rights were bought even before the production process
began, networks seem to play safe by relying more upon the movies that
performed well at the box office counting a somewhat guaranteed television
audience. For example, for the 1992-93 season the three major networks only
bought 23 theatrical movies (CBS 14: NBC 7: ABC 6). But they included such
blockbusters as, Pretty Woman. Total Recall. The Hunt for Red October, Driving
Miss Daisy. Another 48 Hours. Die Hard 2, Ghost and Indiana Jones and the Last
Crusade (Broadcasting. August 3. 1992. p. 21). As Litman (1982) had
predicted a decade ago. the home video. pay television, and other markets
have replaced the theatrical market as the primary transmission outlet for the
movies. The loss of television as a major revenue center. however, was simply
replaced by the newly emerging ancillary markets which now well exceed

theater exhibition in dollars.

23
Box Office: The Primary Window

Until television began showing movies in the mid 1950s, the theater
remained as the only outlet for motion pictures. And the only ancillary income
studios earned came from foreign markets. By broadcasting movies, television
broke the monopoly theaters held as the exclusive window for movies since the
Nickelodeon replaced the Kinetoscope in the beginning of the century. But
neither the industry nor the audiences regarded commercial television as an
alternative window for movies. There is no evidence to show that people
decisively abandoned going to the theater to see a movie knowing that it may
be on television, say, a year or two later. In addition to the long waiting period.
not all movies came to commercial television and the ones that made it to TV
were often edited to keep them within a specific two hour time frame. including
commercials (New York Times, March 9, 1993. p. B3). Therefore, until the
early 19803. with the maturing of premium pay cable. the theater remained as
the primary outlet for movies where most of the revenues were collected and

most of the viewing occurred (Vogel. 1986).

The flourishing new media technologies. such as videocassette. cable.
premium channels. and pay-per-view. whose programming is. for the most
part. built around the movies. however. offered a formidable challenge to the
theaters in the late 19703 and throughout the 19803. By offering un-cut, un-
interrupted movies within a few months after the theatrical releases. these new
outlets were able to redefine the place of the theater in the movie industry. As
theater attendance to VCR rentals reached a ratio of one to four.5 theater is no

longer where most of the viewing occurred. Nor is it where most of the

 

5m 1992 4.1 billion videos were rented against the near one billion movie tickets sold.

24

revenues are collected. As shown in Table 1.3. the box-office share of the
domestic movie revenue now consists of less than a third and it had dropped
10 percent during the period of 1985 and 1990 with a compound annual
growth of a mere six percent. The decline has taken place despite the rise of
the average theater admission price from $3.55 to $4.75 (Quigley. 1992). A
breakdown of worldwide revenues reveals even a weaker position for the
theatrical grosses. In 1980. for example. 80 percent of worldwide revenues
were attributed to the theater grosses. Ten years later. in 1990. only 30
percent of the worldwide revenues came from box-office (Time. July 30. 1990.
p. 56). This. however. did not upset the industry because. even though the
revenue centers have changed. the overall revenues continue to grow.
According to industry analysts, with several sources of revenues available.
large grosses at the box-office are not that crucial in the present economic
equation. In the growing multi-channel marketplace, the role of the theater is
not necessarily to recoup the costs and make profits. Rather, says Variety‘s
Art Murphy:

It [the theater) is the launch pad, and the big hit there is the big hit in all

markets-in videocassettes. in pay. cable. network and syndicated TV.

and all those markets around the world. A fllm’s main job is to establish

itself as something the public wants to consume in the future. where the
real money is. (italics added)

Like Murphy. Douglas Gomery (1993). calls the first-run theater the
principal “voting booth.” According to Gomery. feature films still begin their
marketing life in theaters and the theatrical performance is still crucial to the
consequent ancillary markets. There is empirical evidence to support the
notions of theater as the “launch pad.” or the ”voting booth.” A recent survey

of video store customers. for example. found 60 percent of those surveyed

25
leaving stores without renting anything because the movie they wanted to rent

wasn‘t there. Moreover. the survey also found that people didn‘t choose
alternative tapes because they were unfamiliar with the available options (New
York Times. January 4, 1992. p. B4). So the theatrical release and the
publicity attached to it play an important role in the performance of a movie at
the markets that follow the theatrical release. Theater. therefore, is where the
movie is introduced and publicized. attracts the media attention. gets critiqued
and reviewed. builds the word-of-mouth publicity and eventually writes its own
financial fate. As such. the theater appears to remain as the primary window
in the release hierarchy while its financial importance has given way to newly

emerged ancillary markets.

The Other Windows

As noted elsewhere in this chapter. currently two thirds of the domestic
movie revenues and about 70 percent of worldwide revenues are earned from
sources other than theatrical grosses. The non-theatrical shares continue to
grow as technological advances constantly add new windows to deliver movies.
For example. compared to the nearly 24,000 theater screens. there are 60.000
video rental outlets in the country. The importance of ancillary markets is
such that acquisitions and mergers are planned and implemented with those
markets in mind. For instance. when Disney recently acquired Miramax6 as an
autonomous division of Disney’s distribution arm Bvena Vista Pictures. it '
specifically wanted to distribute eramax's future films and to acquire video.
cable, TV. pay-per-view (PPV) and all ancillary rights to its current library of

over 200 films (New York Times. May 1, 1993. p. 17). The most important

 

6The independent distributor of mostly art films who released such low budget box office
successes as. The Crying Cam. Sex. Lies and Videotape. My Left Foot etc..

26

financial centers of these “other windows,” however. are Videocassettes and
pay-TV (or premium channels). while PPV. video-on-demand (VOND) and laser
discs are making inroads to the market as their share is also on the rise.
Following is an analysis of these new windows and their financial strengths and

effects on the movie industry.

Videocassettes

For every theater in the country that offers a single movie at a given
time. there are two-and-a-half video rental outlets that offer virtually thousands
of movies. The growth of video rental stores. from a mere 2.500 in the 1980
to a phenomenal 60.000 in 1990. shows the rapid growth of the home video
market. The home video market currently stands as the financially most
important outlet for the movies. As observed by the industry investment
bankers Veronis Suhler 81 Associates (1992). no other communication medium
showed a similar growth during the past 15 years than the VCR. According to
most recent estimates. about 80 percent of American households are now
equipped with a VCR and the penetration is projected to reach 88 percent in
1995 (New York Times. February 4. 1993. p. B4: Veronis. 1992). Although
purchase of a VCR is triggered by many of its potential uses. such as time
shifting7 and watching home videos. watching prerecorded videos appears to
be the major use of the VCR. Rental of prerecorded movies currently
surpasses all available means of watching a movie. Although no data are
available for the average audience for a rented video, even a very conservative

estimate of 1.5 per video rental would put the annual audience over 6 billion

 

7 The ability of the viewer to tape a show for later viewing as well as eliminating commercials by
editing them out or fast forwarding them

27

mark. As Figure 1.1 shows. consumers spent nearly $12 billion to rent and

purchase

Figure 1.1

 

 

Consumer Homevideo Spending
1985-1992 In Millions

$1 2,000

 

 

$1 0.000

$8.000

 

 

 

$6.000

 

 

 

 

$4.000

 

 

 

 

$2.000

 

 

 

 

 

 

 

$0
1985 1986 1987 1988 1989 1990 1991 1992

+Sellthrough +Rental +Total

 

 

 

 

Source: International Television and Video Almanac, Quieg Publishing Co., 1992

 

prerecorded video cassettes in 1990 of which over 75 percent consisted of
adult and children feature films (Quigley 1992). Over the period between 1985
and 1990, home video end user spending increased at a 32.7 percent
compound annual rate. making it the fastest growing sector of the

communication industry (Veronis. 1992).

The rental figures closely correspond with the available empirical
evidence on moviegoing behavior. According to a survey done by Alexander &

Associates. an industry consultant firm. 67 percent of the respondents prefer

 

28

to watch movies at home rather than go to a theater while only 22 percent
prefer to go out to a movie (New York Tlrnes. February 6. 1992. p. B5).
Although there is no evidence to show to what extent the theater audience
overlaps with the rental audience. the stagnant nature of the theater
attendance over the past two decades in contrast to the rising rental market

suggests that videos are bringing new audiences to the movies.

Economical. social and technological factors seem to explain the recent
upsurge in the home video market. Economically. it is much cheaper to rent a
movie than go to the theater to see it. Ticket prices have risen 74.4 percent
since 1980 to a current average of $5 as against the average video rental price
for top rental currently range from $2.25 to $2.50 which is. in fact. coming
down as the number of video stores flourish (Dept. of Commerce, 1992).
While one ticket admits one person to the theater. the video could obviously

entertain an entire family thus making the purchase a more economical one.

Home video also eliminates the refreshment expenses which are a major
part of the movie bill. and in many cases the baby-sitter cost as well. Home
video was also able to overcome virtually all the social problems attached to
going to the theater which Samuel Goldwyn and Paul Lazarsfeld saw as reasons
for the deterioration of the movie audiences in the 19503. It eliminates the
logistical problems such as arranging a baby sitter. having to hurry through the
dinner. and driving several miles and finding a place to park. With VCR. one
can also avoid the “distasteful” feeling of leaving the house to see a movie.
Above all, unlike television. which offer a limited number of edited movies with
commercial interruptions. video releases in fact. have begun to offer more than

the theatrical release. For example. some movies are released on video with as

29
many as three different versions with different ratings (R. NC-17, Un-cut). and

some are released with added footage known as the ”director's cut.“ As such.

the video version makes the movie more attractive than the theatrical version.

Technologically. it is now possible to create the theater-like visual and
sound effects at home for a price affordable to the average family. Virtually all
television receivers with screen sizes greater than 20 inches are now equipped
with stereo sound as a standard feature. And according to industry sources.
unit sales of big screen television sets are expected to increase by 30 percent
this year (New York Times. June 6. 1993. p. C1). While it took more than 30
years for television to deliver stereo broadcasting9. home video got a head start
by offering VCRs with stereo sound at an early stage. Over ten percent of the
VCR households are now estimated to have machines with built in Multi
Channel TV Sound stereo broadcasting (MTS) capability and over one million
households with laser disc players which also reproduce superb quality picture
and sound. The increasing popularity of this state of the art home video
equipment, whose primary purpose is to enhance the quality of movie viewing

at home. indicates the seriousness of the home video audience.

The industry data on home video also suggests a growing trend among
the public to own copies of their favorite movies. A third of revenues in the 11
billion video industry came from purchases and the share is growing (New York
Times. May 21. 1992. p. B 4). According to Veronis. Suhler and Associates.

(1992) from 1986 to 1991 sales of pre-recorded videos grew almost 31 percent

 

8Recent such releases include Godfather II! with additional 12 minutes: JFK with additional 7
minutes: and re-release of longer versions of Lawrence of Arabia: The Graduate and Blade
Runner.

9It was in 1984 the FCC approved the Multi Channel TV Sound broadcasting (MTS Stereo)

30

while rentals rose only 13 percent: and the Department of Commerce estimates
that there are about 100,000 business establishments engaged in video
cassette sales (U.S. Industrial Outlook. 1992). As seen in Figure 1.1. sales
accounted for more than 4 billion dollars in 1992 and purchase of videos tends

to grow as rentals appear to stabilize with VCR penetration near saturation.

In fact. it was the introduction of VCR that allowed the average family to
own copies of their favorite movies and consumers seem to accept the notion.
Aladdin, for example, sold 10.5 million copies in the first three days of
release. topping the $217 million it grossed in theaters (Newsweek, October
18,1993. p. 58). According to a Variety survey. during the first two months of
1993 sales of all kinds of tapes were 15 to 20 percent higher than a year ago.
And. as expected. movie titles tend to dominate the retail sales. For example.
of the 19.6 million cassettes sold during the first two months of 1993. about
11 million were movies. The 1993 figures are bound to do better with Beauty
and the Beast selling 20 million copies and Aladdin selling 25 million copies by

the end of the year (Advertising Age Fax. December 16. 1993).

Responding to the consumers' demand. studios are offering more lower
priced titles than ever before. Pricing strategies adopted by the studios show a
quite effective marketing plan that help increase the sales. Depending on the
theatrical performance of the movie. videos are priced either for rent. sale, or
rent first and then sale. Movies that did well in the theaters, i.e. blockbusters
in the $100 million class at the box-office and the movies that appeal to
children, are often priced under $25 with a street price around $15 or less.
Others are targeted towards the rental market and are priced between $75 and

$100. a form of price discrimination. Sometimes a release that is priced for

31

rental is later lowered to a selling price. For example. after receiving all four
major Oscars (Best Picture. Best Director, Best Actor and Best Actress) Orion
re-released Silence of the Lambs under $20 which was originally priced at
$99.98. The same studio sold an estimated five million copies of the following
year’s winner of seven Oscars, Dances With Wolves. within two months priced
at $7.99 in an exclusive deal through McDonalds in 1992. it also had an initial
price tag of $99.98. Although the rock bottom prices Orion offered were an
attempt to come out of its financial difficulties, the marketing strategy worked
so well that the other studios began to follow suit this year. In a recently
announced deal between Paramount and McDonalds. Ghost and Charlotte’s
Web are scheduled for the Summer release (priced $5.99 each). The mass
marketing of a title is also used to promote the release of the sequel as well.
For example. Addam’s Family and Wayne’s World were sold for less than $6 to
promote the theatrical sequels Addam’s Family Values. and Wayne's World II
-to be released in November and December 1993 respectively (New York Times.

May 27. 1993. p. B4).

 

 

 

Table 1.4
UNIT TRANSACTION AND GROWTH PER VCR HOUSEHOLD

Year Purchase Per HH Growth (Percent) Rentals Per HH Growth (Percent)
1985 0.9 — 29.2 —

1986 1.3 44.4 28.8 -1.4

1987 2.0 53.8 33.6 15.1

1988 2.4 20.0 47.4 41.1

1989 2.8 16.7 52.7 11.2

1990 , 3.1 10.7 51.2 -2.8

 

Source: Communication Industry Forecast. July, 1992

32

To recoup the most out of the rental and sales market. the studios and
video distributors are beginning to provide big titles more quickly than in
previous years (New York Times. June 18. 1992. p. B4). Some titles are in fact
released on video while the movie is still running in second run theaters, and
the video release schedules and promotions are being planned with the
theatrical release. As noted elsewhere in this chapter. the video release is
planned to get the benefit of the theatrical promotions by offering the video

before the memories of the movie fade away from people’s minds.

Pay-Per-View

In retrospect, it is evident that the extravagant predictions made a
decade ago about how pay-per-view would revolutionize television and movie
industries and send the video stores out of business by siphoning the home
video consumers did not quite materialize (New York Times. May 24, 1993. p.
BS). After a decade of existence and an impressive 21 million homes10 with
pay-per-view capacity. pay-per-view is still trying to find its place in the home
video industry. Pay-per-view was built on the premise that it can profit from
major television events. such as sports. musical. and prize fights that cost the
viewer somewhere between $15 to $40 per event and from the movies just off
from the theaters that cost the viewer $4 to $8 per request. But the majority
of the events that PPV expected to be profitable flopped as most of the events
failed to get more than one percent of the PPV households requesting the
events. Except for very few prize fights. wrestling and live concerts, the
subscription rate fell well below the industry expectations of the 10 percent

rate (New York Times. November 12, 1992. p. B5). Nor did movies on PPV

 

loWhich exceed the “critical mass” of 20 million households the PPV industry thought in the
beginning it would need to thrive

33

make an effect on the videocassette industry. The greatest impediment to
PPV. according to Dantia Gould. the editor of Pay-Per-View Update. a monthly
trade paper for the industry. is the cost. As Gould has observed:

If the average cable bill is $30 to $35 a month you can see that if

someone orders a fight for $35 and then a concert for $20 and then a

couple of movies at $3.95. you wind up with a cable bill of $100 a

month. I don't know if I‘m really going to watch that much if the bill is

going to be that high.

As a result. however. pay-per-view today is dependent more on offering
movies and less on major events. There is no evidence to show that the
regular video rental clientele is deserting the video stores in favor of PPV in a
harmful magnitude. Rather. both systems appear to co-eidst. However. as a
PPV movie cost two to four dollars more than the video rental. it is unlikely that

PPV offers a threat to the video stores and the cassette industry unless it is

highly price competitive with video stores (Vogel. 1986).

In the meantime. there are some recent developments in the PPV
industry to suggest that it is willing to test some new waters. For example. a
recently announced joint venture by Carolco Pictures (producer of such big
budget blockbusters as Terminator. T2. Total Recall. Basic Instincts. and Rambo)
and Tele-Communications lnc., (country’s largest cable company) revealed TCI's
plan to invest as much as $90 million in Carolco in return for the rights to
show up to four new Carolco movies on its PPV network three times on the
weekend before they are released theatrically. The first of the four films is
expected to be shown late 1994. Although the movies were not yet identified.
it is speculated that a Carolco-Schwarzenegger movie will be one of them.
According to industry analysts. whichever movies are chosen they will have to

be major ones. because the screenings will be sold as ”events” at fees

34
estimated to be between $20 and $40. The home video industry. however.

took the move as a positive one for them. Commenting on the Carolco-TCI
deal. an industry analyst at Paine Webber noted: “For video stores. it will be
like having an extra-wide theatrical release” which will promote the movies for

the video market (New York Times. May 6. 1993. p. B4).

Piracy has also become a problem for pay-per-view more than the other
windows. A recent industry survey revealed that nearly two million households
have illegal descramblers, According to the same survey, conducted for the
Video Software Dealers Association. when viewers were asked where they first
saw hit movies, 25 percent said theaters. 50 percent said on videotape and 25
percent said on pay-per-view. which is much larger than the loyal pay-per-view
subscribers (New York Times. November 29. 1993. p. BIO: October 1. 1993.
p. BS). More than losing revenues. piracy hits hard on pay-per-view as studios
lose faith on the medium. Due to the increasing number of illegal
descramblers, studios now give big titles an exclusive 30-to-35 day window in
video stores before making them available to pay-per-view. Expecting high
cassette orders from video stores, Paramount even extended the video window
to 80 days for three recent hit movies Indecent proposal. Silver and The Firm,
delaying the films' pay-per-view release dates to as late as March 1994 (New

York Times. November 29. 1993. p. B10).

Looking at the sheer differences in consumption, price. and technical
difficulties. industry experts see no immediate effects of pay-per-view on video
rental. In 1992. for example. fewer than one movie per month was ordered by
the pay-per-viewers while video store customers rented an average of 5.6

videos per month (New York Times. November 2. 1993. p. B10). The future of

35
pay-per-view. however. will likely to be determined by the performance of the

latest arrival in the media industry: video-on-demand. in which pay-per-vlew

becomes just one of the many players.

Video-On-Demand

Beyond several localized experimental areas. video-on-demand is from
several years to maybe a decade away until its full potentials are offered (New
York Times March 11. 1993. p. B5). Entertainment is just a part of this latest
technological offering in which telephone. computer and entertainment
industries are forming an alliance to offer a la carte video services for
Americans to shop, bank. learn. communicate and entertain. All these
activities become a part of the “Information Superhighway” that is being
developed under the auspices of the Clinton Administration. According to then
Apple computer chairman John Sculley, the revenues generated by this mega
industry could reach $3.5 trillion worldwide by the year 2001 1‘ (Newsweek,

March 1. 1993. pp. 75-76).

The competition for the dominance or a large share of this supposedly
lucrative venture is already in full swing. All major players in the
entertainment. telephone and computer industries have already made financial
commitments to video-on-demand. For example. TCI is spending $2 billion on
fiber optics and several hundred million dollars on digital-compression
techniques to create a 500-channel cable system. Viacom International. which
operates cable systems with a total of 1.1 million subscribers and owns cable
programmers that include MTV and Showtime movie channel. and AT&T

announced a market test of new technology that delivers movies on demand. as

11The entire U.S. GNP today is about $5.9 trillion

36

well as games and shopping services over cable-television networks. AT81T’s
Bell Laboratories is committed to provide the technology needed for advanced
switches for routing information and large computers for sorting and retrieving
movies and other programs. Just two weeks before the Viacom AT&T deal. US
West announced that it would invest $2.5 billion with Time-Warner to build
advance cable and information networks that would offer services similar to
AT&T/Viacom (New York Times. June 2. 1993. p. Cl). In the meantime
Microsoft. the biggest software maker in the world. is reportedly exploring a
possible alliance with several cable and entertainment companies (including
Time-Warner and Tele-Communications) to establish Microsoft software as the
standard for interactive TV that is being implemented. According to the
industry experts. the potential alliance helps meld computing and television
that would enable viewers to sift through huge amounts of data - from first run
movies to encyclopedia entries to the morning news - using their TVs (Wall

Street Journal. June 14. 1993. p. B6: New York Times June 14. 1993. p. C 6).

As far as video entertainment is concerned. video-on-demand takes pay-
per-view a quantum leap forward. Still cable based. but using fiber optics as
against the conventional copper cables. the new cable system is capable of
carrying virtually thousands of channels. Hence. unlike the PPV, where
customers pick what is available in the limited number of titles being offered at
a given time, video on demand virtually offers. say. all the titles of a major
video store plus all the TV programs. such as yesterday‘s news or last week‘s
episode of Seinfeld by the consumer punching numbers on a telephone or a
television remote control. Although re-cabling the nation‘s 100 million

households with fiber optic cables is estimated to cost billions of dollars and

37
the remainder of the century, the process has already started. In fact. Time-

Warner and TCI are already offering video on demand on an experimental basis
in many states. Since July 1992. 'ICI's true video on demand allows
customers to order from 1700 film titles or other programs any time they want
by pushing a button. Within 10 minutes the movie starts and the fees range
from 99 cents to $4 per film (New York Times. March 11. 1993. p. C1). As can
be seen from the current offerings, expectations and projections, movies. old
and new. represent a major portion of the video-on-demand menu. As such,
the movie industry is in a win-win situation with video-on-demand. If all the
projections were at least partially accomplished. it will greatly increase the
demand for old and new movies. If it failed. there will be no major losses to
the movie industry because its involvement in the venture is only to provide the

software which is already being produced for other windows anyway.

Laser Disks

More than a decade after it was first introduced. and rejected.
consumers now seem to be accepting the laser disc format. Although only an
estimated one million households12 are currently equipped with laser disk
players. the industry believes that the format is growing steadily. According to
the Laser Disk Association statistics. 10 million video discs and 206,000 video
disk players were sold in 1991 compared to the 6 million discs and 159,000

players sold in the previous year.

Several reasons appear to be behind the popularity of the laser disc
format that weren’t there when it was first introduced in the early 19803 by

RCA and the Phillips Consumer Electronics. First. the consumer now appears

 

12Approximately one percent of American households. as against the nearly 80 percent VCR
households.

38

to be more serious about the quality of home video than he did a decade ago.
For the discriminating consumer, laser disc provides superior quality home
video. As Douglas Pratt. the editor of Laser Disk Newsletter. observed: “People
are looking for the next step up in quality” (New York Times. June 3. 1993. p.
B5). Second, there is a reasonably large inventory of movie titles available now
on disc format which justify the purchase of hardware. For example. more
than 10,000 titles are currently available on laser format and virtually
hundreds of new and old titles are added to the inventory every month.

Except for art films and a few independent releases, all movies are now
simultaneously released on both. cassette and disc format. Third, the prices of
the laser disc players are within the reach of the average household. The price
came down from its early 1980 unit price of $1000 to current price of around
$400. For that price a consumer gets a machine that not only plays laser
movies but accepts three inch and five inch format audio compact discs as
well. In a latest development that expects to boost the popularity of the laser
format, two companies, 3DO of California and Philips Consumer Electronics.
announced the release of machines that play back 5-inch compact discs
instead of current 12-inch laser discs. audio CDs and video games. The 5-lnch
CD-ROM videos are expect to lower the production cost allowing companies to
sell a laser movie for about $25 or less. The new format was well received by
the industry which believes the multi-purpose machine will help boost the
consumer acceptance of the laser format (New York Times May 27. 1993. p.
B4). To gain consumer confidence to 5-inch compact disc format. Philips
Electronics has recently announced its decision to release 50 Paramount films

on compact disc format (New York Times. September 24. 1993. p. B4).

39
Compared with the other windows through which movies are delivered.

the laser disc format has some unique features. First. at least for the time
being. it is primarily dedicated to movies. Unlike the videocassette recorder.
pay-per-view or video-on-demand. the purchase of a laser disk player shows a
commitment on the consumer's part to rent or buy movies since the machine's
exclusive or primary function is playing laser discs. With the popularity of
laser discs. the movie industry also foresees a growing market to sell movies.
Unlike videocassettes. all laser disc movie titles are priced to sell. A random
week‘s video releases of the following titles show that while the Videocassettes
were priced at $95. the comparable laser discs were priced between $35 and
$40, The Distinguished Gentleman: Trespass: Hero: The Public Eye: Reservoir
Dogs; The Player: Mr. Baseball. As the current evidence suggests, the
popularity of disc players will eventually provide another window for the movie

industry.

Foreign Markets

Entertainment is America's second biggest net export. trailing only
behind the aerospace industry. According to MPAA estimates the U.S. motion
picture industry earned foreign revenues of $6.35 billion in 1990. an increase
of about 20 percent over the 1989 earnings. that created a trade surplus of
more than $5 billion. This revenue included theatrical exhibition. television.
home video and pay-TV. Movies. however. represent a robust portion of the
entertainment industry and since 1985 the overseas billings for American made
movies have doubled (Corlis. 1990). It is estimated that the U.S. majors
control 80 percent of the worldwide theatrical business and have a majority

share of the box ofiice in virtually every country in which U.S. films are shown

40

(Sochay and Litman. 1992). In Europe. for example. 70 percent of box office
receipts are for American films and one half of India's film imports come from
the U.S.. and in Peru 90 percent of all films shown are American (Bernstein.

1990).

In France, Germany and Italy the U.S. now accounts for two-thirds of all
movie tickets sold (Time. July 19. 1993. p. 67). The top five countries (Japan.
Germany. France. Canada and the United Kingdom and Ireland) accounted for
$692 million of all export rentals (U.S. Dept. of Commerce. 1992). The second
major market. however. is the Far East and for the sixth straight year the
leading export market was Japan with $236.7 million in film rentals for 1990.
an increase of 17 percent compared to 1989. America‘s market share of
movies in Japan in 1993 is 49.7 percent. up from 30 percent in 1988. As a
result of intensive lobbying by MPEAA and U.S. government pressure. South
Korea opened up the market for American movies in the late 19803 and now it
stands as the second largest market for the U.S. distributors in the Far East
(Sochay and Litman. 1992). Practically all U.S. movies now collect more
revenues at the foreign box offices than at the domestic box office. As Table
1.5 shows. each of the 10 top grossing movies in 1992 earned more money

abroad than at the domestic box office.

Foreign markets, like other ancillary markets. are crucial to the survival
of the studios because movies are getting more and more expensive to make.
thus hard to earn profits without the ancillary markets. The average
production costs now amount to $28 million with another $11 million to
advertise and publicize it. The average negative costs have nearly tripled over

the last ten years (Sochay and Litman. 1992). As such, 8 out of 10 movies do

41

not recover their investment costs from exhibition in U.S. theaters.
Consequently. without the financial help of foreign and other ancillary markets

very few movies would even be able to break even at the domestic box office.

 

 

 

TABLE 1.5

10 TOP CROSSING MOVIES - 1992

DOMESTIC GROSS FOREIGN GROSS

TITLE (MILLIONS) (MILLIONS)
BASIC le‘I'INCT $117 $235
HOOK $120 $180
ALIEN 3 $55 $125
JFK $70 $120
PATRIOT GAMES $83
BEETHOVAN $57 $85
FAR ANDAWAY $59
MY GIRL $58 $65
UNIVERSAL SOLDIER $36 $59
FINAL ANALYSIS $29 $46

 

SOURCE Variety - January 4, 1993 p. 5

While the large domestic market and the deep pockets of the
conglomerates help the U.S. film industry to keep the foreign competitors at a
distance. it is the public good nature of motion pictures that allow the studios
to make profits from international markets where the rentals simply exceed the
distribution costs. As a public good. once produced. a movie can be
distributed to an unlimited number of markets and the marginal cost of serving
an additional market is only limited to the distribution cost and the cost of
making an additional print of the negative. As such. it is the first print which
takes virtually the entire production cost. The cost of making a second, third

or hundredth print is just the cost of the print. about $2000 in today's rates.

42
which is negligible compared to the average of $21 million it takes to make the

first copy. International syndicators even get away with these expenses by
“bicycling” (i.e. the same print of a movie is shipped to the next country by the
local distributor as a part of the purchase agreement). Therefore. as Thomas
Guback (1969) succinctly put it. “we can have our films and foreigners can
have them too.” The public good nature of movies thus allows U.S. studios to
function by the principle described by an industry spokesman as, “draw
whatever price a market can bear” (Variety. April 27 . 1988. p 3). As Acheson
et al (1989) pointed out. regardless of the production cost. the studios will
profit by showing a movie in any market in which rentals exceed the costs of

distribution.

While the foreign share of the theatrical revenues now exceeds the
domestic box office share and the total foreign revenues continue to grow, the
revenue structure has changed over the past several years. For instance.
according to Veronis. Suhler 81 Associates. while foreign box-office rentals have
risen 15.7 percent annually since 1985. reaching $1.8 billion in 1990. the
share of the average U.S. movie's foreign revenues generated by its theatrical
run fell from 53 percent in 1985 to 37 percent in 1990. However, as in the
U.S.. home video abroad has become the largest source of movie revenues
making up 46 percent of the foreign total. or $2.3 billion. In 1985. by
comparison. home video accounted for only 33 percent of foreign movie
revenues (Table 1.6). Home video has been the fastest-growing revenue
stream in both domestic and foreign markets. Hence. as Sochay and Litman
(1992) pointed out. new technologies, as in the historic past. have helped

rather than hindered economic opportunities.

43

 

 

 

 

TABLE 1.6
FOREIGN REVENUE AND SHARES FOR MOVIES
1985 1990 Compound

Category (5 Millions) Share (5 Millions) Share m
Box Office Rentals 887 53% 1,836 37% 15.7%
Television 237 14% 863 17% 29.5%
Videocassettes 564 33% 2,305 46% 32.5%

Total 1,688 5,004 24.3%

 

Sources: Veronis, Suhler 6t Assoc, Paul Kagan Assoc, Wilkofsky Gruen Assoc,

U.S. success in exporting filmed entertainment in recent years is mainly
due to the proliferation of traditional and non-traditional windows in foreign
markets and the increasing privatization of media outlets in Europe. Far East
and in the Third World. The number of theaters in Europe and Asia are in the
rise. For example. new theaters have been built or are being planned for
France, Germany, Italy. Spain. and Australia (Sochay and Litman. 1992). Most
of these new theaters in Europe are joint ventures led by American investors.
In Great Britain. for example. modern multiplexes have added 350 new
screens that had doubled the theater attendance over a six-year period (New

York Times. March 7. 1990. p. C1).

The rapid growth of foreign markets also affect the form, marketing. and
distribution of American films influencing which movies get made and who
stars in them. Movies that convey visual humor. action and special effects or
contain little dialogue typically travel well overseas. For example. Warner
Brothers made the third. fourth and fifth sequels (and now the sixth). to ”Police

Academy” partly because the first two did well abroad. even though American

44
interest seemed to wane after the first (Variety. September 9. 1991. p.1: New

York Times. June 25, 1990. p. A1). So is the appeal of certain stars overseas.
According to New York Times movie analyst Geraldine Fabricant. Carolco
Pictures was able to raise about $60 million to make “Total Recall.” because it
13 an action film and Arnold Schwarzenegger is an international star. Similarly.
Warner Brothers had no qualms about producing ”Lethal Weapon” or its sequel.
both starring Mel Gibson. Yet. as no single overseas market is bigger than the

domestic U.S. market. Hollywood executives still base their decisions primarily

on the chances a film has in the U.S.

The success of American movies overseas. however. has come amidst
variety of trade barriers foreign governments employ to restrict foreign. mainly
U.S. , film into national markets and growing piracy especially in the Third
World (Sochay and Litman, 1992: U.S. Dept. of Commerce, 1992). A growing
concern is quotas. which essentially limits screen time in theaters and
television broadcast time and also restrict import of foreign films. To
strengtl'ien the local movie industries against the foreign movies (especially
American). many foreign governments also offer subsidies for production and
distribution of national productions. place restrictions on foreign remittances
and turn a blind eye on copyright violations. According to U.S. government
sources trade barriers and reasons for erecting them vary from country to
country. Some foreign governments want a greater share of the profits while
the Others seek to protect indigenous film industries or to defend their
language and culture against a perceived threat of foreign cultures (U.S.

Indus trial Outlook 1992). Whatever the objectives. the trade barriers reduce

export earnings for the U.S. film industry and according to a senior

45
administrative official the U.S. take “would be significantly higher without the

barriers” (Time. July 19. 1993. p. 67). The issue became so crucial in foreign
trade negotiations that the most recent GATT (General Agreement on Tariffs
and Trade) talks came to a deadlock on the issue of filmed entertainment. At
present GATT or E.C. (European Community) do not have specific screen
quotas on foreign (non-E.C.) movies. However. for TV . the E.C. now requires
that 50 percent of each nation’s programming be produced in the E.C. France.
in fact. went further by requiring that 40 percent of their TV shows be

produced in France.

The specific regulations. trade barriers or nontariff barriers that affect
the movie industry is yet to be seen. The tariff and nontariff barriers.
however. are not new phenomena and had been tried by European and other
nations for over half a century to protect indigenous film industries and
perceived threats to cultural sovereignty. As such. their true impact beyond
the rhetorics should be evaluated in historical perspective. For example.
Britain. Germany. France and Canada all had enacted some kind of laws that
l'mluirecl film distributors and exhibitors to show minimum quotas of local fare.
Yet. the U.S. movie industry has historically been quite successful in
Penetrating into all those markets. One very successful step taken by many
Meagan film-makers was to produce low-quality films, known as “quota
quicmes». so that the more financially rewarding American films were qualified
to be shown. As Acheson et. al. (1989) pointed out. “many of those domestic
films Were of such poor quality that they were never shown. but they were
necessary to allow an imported film to qualify for exhibition.” Nrthermore. as

Sochay and Litman (1992) pointed out. “the stability of a monolithic European

46

cartel is still questionable. Despite the move toward a single market. there is
no such thing as European cinema. Rather it is more realistic to think of it in
terrns of French. Italian. German. and British cinema.” Moreover, only the
Hollywood inventories will be able to fulfill the demand created by increasing

windows opening up in E.C. nations and the Central and Eastern European

cou ntries .

While Hollywood won eventual copyright protection for its movies
evel'ywhere in the world (a victory. if properly implemented. that could add
billions to Hollywood studios as MPAA estimates that piracy causes revenue
losses in the range of $1 to 1.2 billion per year). in GATT and E.C. talks trade
issues like the export of U.S. entertainment programs to Europe were shelved.
The U.S. also failed to eliminate European and other domestic subsidies given
to tl'le U.S. competitors (Newsweek. December 27. 1993). Despite the
regulatory difficulties the U.S. film industry appear to be in a very favorable
situation in the global market. First. the expansion of television systems and
the p0pularity of home video and other technologies throughout the world is
bound to increase the demand for U.S. filmed entertainment products as only
the U-S. majors are currently in a competitive position to meet the global
demmd. Second. the increasing production costs for movies seems to favor

the U-S. film industry with its large domestic market and the deep pockets of

the col'lglomerates.

47

Summary

The preceding discussion shows that the movie industry is not in the
demise but holding quite strongly its place in the entertainment industry while
the growing ancillary markets continue to support it. Although the traditional
window. the theater. is in the decline. the audience for movies. in general. is
on the rise perhaps unprecedented to any historical time. For example. while
the best years for movies put the average weekly audience at 90 million in the
late 1940s. a very conservative estimate of the current theater attendance and
video rental figures alone put the weekly audience size at least in the vicinity of

100 million.

It is also important to recognize the impact of technology on the film
industry. As seen in the preceding discussion. the movie industry is primarily

technolog driven. As Vogel (1986) has observed:

Unquestionably the most potent impetus for change over the long term
has been. and probably will continue to be. the development of
technology. In the filmmaking process itself. for instance. the impact of
technological improvements has been phenomenal. To see how far we
have come. we need only remember that “talkies” were the special-
effects movies of the late 1920s.

Undoubtedly. it is the technology that made Jurassic Park the highest
grossing movie of film history with over $1 billion world wide rental revenues
and. Historically. Vitascope came to rescue Kinetoscope and the introduction
of sound brought the audiences back to the theaters. When a competing
technology-television-diverted the audience’s attention. color. 3D.
CinemaScope. Stereophonic sound and other technological breakthroughs
helped the survival of the industry until the competing technology itself became
the savior. The financial benefits of the home video to the movie industry is

unparalleled to any ancillary market the industry has seen in the past. Other

48

technological developments. such as pay-per-view. video-on-demand. and laser
disc. are there to capture any fallout from videocassette users. Hence. the
theater is neither dead nor dying. While its financial significance continue to
weaken as the other windows become the new financial centers. theater seems
to remain. as Murphy called. the “launch pad” where movies are introduced to
the public. Moreover. as some industry experts concede. the growth of
ancillary markets may also be responsible for brining new audiences back to
the theaters. As Marc Merry. AMC Theaters director of corporate advertising
noted on the recent success of the box-office:

Unquestionably ancillary revenues are behind the increase in the

number of films being made. But I also believe the general quality of

production has risen and that the video business is responsible for more
people going out to the movies. I‘m certain that a lot of people saw

“Terminator” on tape and it made them want to see the sequel in a

theater (Variety. November 29. 1993. p. 1)

Above all. the technoloy has literally liberated the viewer. The
unprecedented access the viewer now has to filmed entertainment. that is. his
ability control what. when and where to see cannot be easily disregarded when
analyzing the movie industry economics and its audiences. The importance of
all theses windows to the audiences and their impact on movie attendance will

be the focus of the remainder of this research.

Chapter 2

Motion Picture Attendance Research: A Review

Introduction

Considering its long existence. the movie audiences remarkably have
been the least studied in mass media research. The industry did not see any
compelling reasons to study the audiences until the new media. especially
television. began to have a direct effect on movie attendance. The first known
attempt by the movie industry to systematically study the audience reportedly
took place in 1957 when the Motion Picture Association of America (MPAA)
commissioned a survey to examine the moviegoing behavior among Americans
(Austin. 1991). The newfound interest in studying the audience. in fact, came
when television and other alternative leisure activities effectively claimed the
consumer‘s leisure dollar and time. The findings of subsequent audience
research carried out by the industry. however, remain proprietary (Austin,

1981).

While the industry attempted to understand audience behavior with a
commercial interest and kept the findings to themselves, academic researchers
concentrated more on the aesthetic aspects of the medium than its audiences

(Austin. 1981. 1991; Dominick. 1983). Only over the last ten to fifteen years.

49

50

did an emergence of serious interest surface among media scholars to study
movie audiences from a communication and economic perspective (Austin.
1981. 1981b. 1982. 1984. 1984b. 1991: Dominick. 1987; Litman. 1979.
1982. 1983: Litman & Kohl. 1989; Musun. 1969). Yet. they fell far behind the
amount of audience studies done on. say. television. radio or newspapers. The
lack of such research. argues Litman and Kohl (1989). “is largely traceable to
the proprietary nature of financial data in the industry and the fact that certain
creative/artistic elements in filmmaking are so complex (and subtle) that they
remain beyond even the most sophisticated content analysis attempts to

quantify them.”

Academic research on movie attendance. however. concentrated mainly
on two broad but essential research questions: (1) Why people choose to go to
movies from among the wide array of entertainment options? And. (2) Why and
howa particular movie is selected over all the other movies available at a given
time and place? There are two dominant research approaches that sought to
answer these questions. One is identified as the communication theory
approach and the other as the economic/business approach (Litman. 1991).
The communication theory approach uses social/psychological theories to
explain why people choose to go to movies and why a particular movie is
selected. This approach looks at the individual motives for movie attendance.
The economic/ business approach. on the other hand. studies the economics of
the industry to explain movie attendance among a group of people as a .
precursor to financial success of the movies. This line of research looks at
various elements of the supply side of the industry. While it acknowledges the

effects of such demand factors as price. income. quality of the product and

51

available substitutes on the moviegoing habit of groups of people. the
economic/ business approach primarily focuses on the institutional factors
affecting the entire vertical system of supply of films (Litman. 1989). As such.
the economic/ business approach has concentrated largely on the question why
a particular movie is chosen over the other movies than why people select
movies as an entertainment option. Interestingly. both approaches test the
basic concept of movie consumption (which is. in fact. demand in action) and
share many of the same variables in their analysis. But they differ on the data
sources and manner in which they test hypotheses. The communication
approach uses attitudinal survey research. whereas the economic approach
examines the forces of the marketplace and uses aggregate data from industry

and trade sources (Litman. 1989).

As such. the goal of this chapter is threefold: First it examines the
communication approach to movie attendance and the social psychological
theories behind it. The chapter then examines the economic/ business
approach and its theoretical base. Finally. it attempts to converge the two
approaches to find common ground. theoretically and methodologically. to
explain the movie attendance behavior. Further. an attempt is made to
incorporate the structural and institutional changes taking place in the movie

industry to assess their impact on attendance.

The Communication Research Approach to Movie Attendance

What is now known as the communication theory approach. in fact. is a
two step attempt to explain movie attendance. First, it focuses on why people
choose to .go to movies among all the available leisure/entertainment options

and then asks why and how a particular movie is selected over the others.

52
Although this dissertation research specifically focuses only on the latter. it will

nevertheless introduce the necessary literature pertinent to the question of

“why go to the movies” as a prelude to the present study.

Why Go to the Movies?

Most of the research that looked into people's motives for attending
movies attributes such desires to relaxation. entertainment and to learning
(Austin. 1981. 1984. 1991). The audiences. it has been argued. willftu select
media content to satisfy social. psychological and intellectual needs. which are
also the motives of leisure in general. Socially. movies serve as a lubricant.
That is. movies may facilitate social contact and discourse in situations far
removed from the theater. Psychologically. movies offer escape from problems
and boredom. As such. attending movies. argues Austin. provides an
experience similar to “attaining the psychic equilibrium deemed so valuable by
many psychological theorists.” lntellectually. movies help experience and
understand things outside people's immediate surroundings. That is. movies
fulfill intellectual or educational needs beyond their predominantly
entertainment orientation. Accordingly. movies make people's own world more
bearable. satisfy intellectual. aesthetic and religious needs. and provide
relaxation. Two dominant social/psychological theories. the expectancy-value
theory of motivation and the uses and gatifications theory. provides the
theoretical framework necessary to explain how audiences satisfy those innate
needs through movies. What follows is a brief review of those theories. their
applications and a critique of the general applicability of them to explain movie

attendance behavior.

53
Expectancy-Value Theory and Its Applications to Moviegoing

The expectancy-value theory. or rather theories. stem mostly from the
early works of Tolman (1932: the expectancy concept). Lavin (1935: the
valence concept). Atkinson (1964: achievement motivation theory). and Vroom
(1964: work motivation theory). The expectancy-value approach to behavior is
governed by two components. expectancy and value. The expectancy
component refers to beliefs that actions will lead to certain outcomes (i.e.
gratifications). and the value component refers to the value of these outcomes
(i.e. the concept of valence). The theory also holds that the expectancies and
values of outcomes combine multiplicatively to determine subsequent behavior
(Galloway and Meek, 1981). As such. if E1 is the expectancy that an action
will lead to outcome i. and V1 is the value of this ith outcome. then the overall
value for an action is given by 2.5V.- . The action chosen would be the one
that maximizes the expectancy-value summation (Bettman. 1979). When
applied to consumption. the consumer takes an action to select a specific
option from a relatively homogeneous class of events that maximizes the

expectancy-value summation.

Applying the expectancy-value theory to explain why people attend
movies. Austin (1991)13 regards moviegoing as one in a cluster of several
(presumably homogeneous) leisure options. For a given option the consumer
assigns a weight based on his or her subjective evaluation of the outcome's
importance or desirability and generates an outcome expectancy term. which is

the consumer's best guess about the probability that choosing each leisure

 

13A search of literature into this area found no other attempts to apply expectancy-
value approach to moviegoing. There were. however. several applications of the
expectancy-value theory to explain many aspects of television viewing behavior which
will be referred to later in this chapter.

54

activity will generate that specific outcome. The process also assumes that the
consumer mathematically calculates the outcome of each leisure activity and
the comparisons of each outcome result in the choice that maximizes the
expectancy-value summation. However. given the theory’s “general
applicability.” Austin (1991) shows some reluctance to use the expectancy-value
approach exclusively to explain the movie selection process. Instead. he favors
the uses and gratifications approach which focuses specifically on the mass

media to explain movie attendance.

Uses and Gratifications Approach to Movie Attendance

The uses and gratifications approach has been the dominant paradigm in
media research during the past two decades. As Austin (1991) observed. the
uses and gratifications approach indeed focuses on mass media. In fact. a
large majority of mass communication research carried out during that period
has roots in the uses and gratifications approachs. more than any other single
theoretical framework. As Palmgreen (1984) has noted “since the publication of
Blumer and Katz’s (1974) landmark volume. The Uses of Mass Communication.
research into the uses audience members make of the mass media. the
gratifications derived from media consumption. and their antecedents and
consequences has continued at an accelerating pace.” The uses and
gratifications approach departs from “direct effects” and “limited effects”
research and claims to look at what people do with media rather than how

media affects their lives.

Although its roots can be traced back to some earlier works. the origin
of gratifications research is attributed to Elihu Katz's writings in the mid 1950's

where the now widely quoted statement “ask not what the media do to people.

55

but what people do with media” was derived. Contrary to the passive role
assigned to the audiences by the “effects” research. the uses and gratifications
theory postulates that: The audience is active. conscious and aware of their
interests and motives: their media use is needs-and goal-oriented: they bend
the media to their needs more so than they are overpowered by media: and
media compete with other sources for need satisfaction (Katz. Bulrner. and

Gurevitch. 1973: Elliott and Quattlebaum. 1979).

Austin (1991) notices conceptual similarities between the expectancy-
value theory and uses and the gratifications approach. To Austin. the uses and
gratifications approach serves as a substitute theoretical framework for the
expectancy-value theory to explain movie attendance. “Uses.” argues Austin.
are similar to that of motivation while “gratifications” resembles the expectancy-
value concept. Similar to the expectancy-value theory. the gratifications
approach relies on the concepts of outcome values and outcome expectations.
As the gratifications approach assumes an active audience. they purposefully
select and attend to media to satisfy various needs. The rational (active)
consumer’s use of movies. therefore. is said to be aimed at gratifying his or her
social. psychological and intellectual needs. Since audiences gain familiarity
with media over time. argues Austin. “prior media experience helps determine
media use as a means to fulfill needs.” This indicates. as Weibull (1985) has
noted. that uses and gratifications studies have focused excessively on habitual
media behaviors. While both these approaches show some conceptual

promises. their theoretical assumptions fall short in explaining “moviegoing.”

56

Shortcomings in Expectancy-Value Theory and Uses and Gratifications
Approach to Movie Attendance

As Palmgreen and Rayburn (1979) had observed [with reference to the
uses and gratifications approach). “the true test of any theory lies in its ability
to explain (and therefore predict) certain phenomena.” Nevertheless. despite
their many contributions. expectancy-value theory and the uses and
gratifications approach fall short of being a formidable theoretical framework
for explaining. and consequently predicting. movie attendance behavior.
Although the uses and gratifications approach theoretically claims to have a
general applicability across media for a wide variety of situations. attempts to
test empirical associations between various gratification measures and media
exposure. medium choice. and content choice have overwhelmingly been
limited to the medium of television. In a review that was similar to a stock-
taking of the uses and gratifications approach research. Palmgreen. one of the
prominent advocates of the gratifications approach himself. admits the
television bias in gratifications research. According to Palmgreen (1984). “a
majority of studies to this point [until the mid 1980s) concern television.
including total TV exposure. exposure to different program content types such
as. television news. debates. and quiz shows. public television viewing. and
viewing of specific programs.” When the empirical applications of the
gratification research deviate from television. it is often limited to newspapers
and radio (Nordland. 1978: Rosengren. and Windahl. 1972: Kline. Miller. and
Morrison. 1974: Becker and Fruit. 1982). These applications legitimize
Weibull's (1985) criticism that uses and gratifications research has focused
excessively on habitual media behaviors. Unlike television. radio. and

newspapers. movie attendance is determined by choice rather than habit. in

57

fact. the idea of choice and selectivity in movie attendance is supported by
Austin’s (1981b) own research where he found that over 70 percent of both
occasional and frequent moviegoers indicated that they were most likely to go

to a movie when a picture of particular interest to them was being shown.

Only recently. Palmgreen and others (1987) attempted to explore the
factors affecting the consumption of movies from a uses and gratifications
perspective. In a rather limited application of the gratifications approach. they
found that college students attend movies for a variety of social and information
seeking reasons as well as entertainment. However. as LaRose and Atkin
(1991) noted. “no attempt was made to relate these motivations to movie
attendance behavior or to the consumption of movies in other distribution
channels.” which are becoming quite crucial to the audiences as well as to the

industry.

The lack of movie oriented applications of the uses and gratifications
approach may be partly due to the difi‘iculties associated with conceptualizing
and operationalizing the process. Evidently. the constructs associated with the
uses of. and gratifications sought and obtained from. say. exposure to
television and/ or different program types (such as news. debates. quiz shows
or public television). appears relatively easier to conceptualize and
operationaiize than the constructs associated with motion picture attendance

where choice not behavior determine the use.

Moreover. because the broader research question here is movie
attendance and not selecting a movie. the research task provides a greater
conceptualization and operationalization challenge due to the fact that every

movie can qualify to be a totally new experience. As such. audiences also face

58

the difficulty of seeking gratifications from a totally new experience while
researchers face the difficulty of measuring something “unknown” to the
audiences prior to the experience. As a result. studies that attempt to explain
the movie attendance behavior using the gratifications approach are scarce
among the wealth of gratifications research. The television bias in gratifications
research. therefore. is somewhat tantamount to the drunkard searching under
a street light for his house key. which he had dropped some distance away.

Asked why he didn‘t look where he had dropped it. he replied. “It’s lighter

here!”

Notwithstanding the applicability to the movies. some researchers
attribute the lack of general predictive ability in the gratifications approach to
the absence of a “process” view and inadequacy of an explicit focus on
“expectations.” This urged them to integrate uses and gratifications with
expectancy-value theory to predict exposure to mass media (Galloway and
Meek. 1981: Palmgreen and Rayburn. 1982: Van Leuvan. 1981: Barrow and
Swanson. 1988). Such integrations. however. had not been applied to movie
attendance research. Moreover. other researchers found integrations of

eJCpectancy-value as not good behavioral predictors (LaRose and Atkin. 1991)

As noted elsewhere in this chapter. the application of expectancy-value
tlfileory to explain movie attendance assumes the availability of relatively
homogeneous class of leisure/entertainment options to the potential moviegoer.
hat is. the alternatives are close substitutes for each other. Given the nature
Qf the movie product. this is a somewhat rigorous assumption to meet because
Comparable leisure pursuits to movies may include such option as. plays.

opera. television. concerts. ballet or sports events. The problem with these

ahernz
comm
‘expec
ofhon
consu
outc01
practi
heurh

those

howev
broad
appro;
Where
may d
behav
appro
then:
Mace.
"mow
hasit
Panc

prQCec

59
alternatives is that they demand quite different social. economical and cultural

commitment from the consumer. which might add up to different
“expectancies” and “values.” The difficulty. however. is not only with the lack
of homogeneous product classes but also with the assumption that the
consumer is expected to have some prior knowledge about the potential
outcome of seeing a movie to compare that with the other alternatives. In

practice. however. as Bettman (1979) noted. consumers use a variety of simple
heuristics in making choices and expectancy-value models fail to consider

those heuristics used in making choices.

The most important question arising from this review of literature.
however. is not whether the communication approach is able to answer the
broad research question (i.e. why go to movies). rather. how valid or

appropriate that question is. Under the current structure of the movie industry
Where the majority of the viewing occurs outside the traditional theater. one
may doubt the need of two separate questions to explain the movie attendance
behavior. First. the time order of events proposed by the communication
approach (i.e. first resolving the problem of whether to see the movies and
then what movie to see) may not necessarily be how movie attendance takes
place. There can be many instances where a particular movie triggers the
""1‘ljoviegoing.”l4 an event stimulates interest in itself. As Austin (1981b) himself
h as found. moviegoers are most likely to go to a movie when a picture of
particular interest to them is being shown. As such. the second question may

precede the first or simply nullify the need of asking such a question because

¥

1 4A‘histin himself admits this when he attempts to answer the second question: What
OVie to see?

60

once we are able to answer why a particular movie is selected it automatically

answers the first.
Second. given the question “why go to movies" is a valid one, it overlooks
some important situational factors. such as the roles advertising and marketing
play in attracting audiences and the effects of changing demographics of the
society on movie attendance. Empirical research on the effects of advertising
and marketing on movie selection have found people rely on several information
sources which are readily available in movie selection process (Faber and
O’Guinn.1984). Conventional wisdom also shows the important role advertising
and promotion play in attracting audiences. The advertising budget of a typical
movie. for example. currently run about $1 1 million and that is about 40
percent of the production cost. Besides direct advertising and release
strategies. studios also make use of many promotional tools available to them.
During the first couple of weeks of a new release. stars can be often seen on
'I‘V talk show circles. Tie-in promotions in collaboration with other products to

promote movies and sell-through videos have also become a common practice
in recent years (Maslin. 1993: Magiera. 1992).

Another drawback associated with the communication approach is that it
is more or less based on the assumptions of “moviegoing” prior to the
c<>nditions associated with home video revolution. As seen in the preceding

<2l‘lapter. movie attendance dropped drastically with the advent of television and
with the aging of the population. When the young (the largest proportion of the

movie audience) settled down and started a family. the economics and social
obligations attached to going to the movies became more demanding. That is.

finding a baby sitter. driving to the theater and finding parking as well as

61

dining out as part of the movie outing. made it an event associated with a cost
and many social obligations. It is in this context that television provided
somewhat similar experience “free” of such complications. Until the advent of
home video. movies had to find ways and means to convince the audiences that
going to the movies is something worth getting out of the house for. And until
the VCR and services such as pay-per-view and premium channels came along.
audiences practically had no control over when and where to watch a movie.
and “moviegoing” actually required “going” to the theater. But it is no longer
the case. Now people can create their own movie schedules and evidence
shows that they are increasingly doing so. It is these structural changes that
have led people to rent more than five billion videos last year as against the one
billion theater tickets they bought. Evidently, these structural changes have
not been taken into account by the communication approach in its attempt to

explain the movie attendance. As such. the communication theory approach

faces the difficulty of explaining movie attendance strictly in social.

psychological and intellectual terms.

Why A Particular Movie?
As noted in the preceding section. the communication approach look at

I‘Iloviegoing as a two step process and asks two questions. First. the decision
to go to movies and then selecting a movie. Once the consumer has
rationalized that going to the movies best fulfills his or her needs. what factors
influence the selection of a movie? That is. why go to the movie ‘X‘ rather than
to the movie ‘Y‘? The communication theory approach uses diffusion of
1Grilliovation theory to build a model that purports to explain the process involved

in selection of a particular movie among all the movies available at a given time.

62

The diffusion of innovation theory explains how new ideas. practices or
products are adopted. While the diffusion theory has originated in sociology
and anthropolog'. its applications can be seen in many disciplines. It has
developed across many fields of research over the past several decades and
has been used to explain such diverse innovations as family planning. water
boiling and diffusion of new technological breakthroughs (Rogers, 1983).
Marketing and consumer behavior researchers have also adopted the general
model of diffusion theory for use in their fields to explain new product
acceptance and diffusion over time (Lowrey. 1991). According to Rogers
(1983). who has formalized and popularized the diffusion theory. an individual
proceeds through five steps in the process of adopting an innovation. in the
diffusion process. people first become aware of the innovation and then form
their own judgment about the innovation (these two steps are known as
knowledge and persuasion stages). This is followed by the decision making
stage where the potential adopter weighs the pros and cons of the innovation
and makes a judgment on whether or not the innovation is to be adopted. The
last two steps. implementation and confirmation. occur when the innovation is

actually adopted. An important factor in diffusion theory is time. The time
factor in diffusion research implies that not all people adopt a new idea
slinultaneously. Rogers (1983) identified five adopter categories of
i~l‘1novativeness and use the areas under the normal frequency distribution
<=‘l_1rve to explain their distribution. The adopter categories are: Innovators
( 2 -596): Early adopters (13.5%): Early majority (34%): Late majority (34%) and
Laggards (16%) (Lowery. 1991: Rogers. 1983). When these areas under the

normal curve are stated cumulatively. the logistic curve takes the shape of the

63
popular “S” curve. Rogers also describes dominant personality characteristics

of each of these categories that affect their decision making.

Austin (1991) asserts that. although movies in general can hardly be
claimed as an innovation. any new movie that enters the marketplace can be
considered as an innovation and adoption of that innovation is equal to the
attendance of that film. He then tries to fit the five steps of the diffusion
theory to movie attendance by explaining the steps involved in attending a
movie from first hearing or reading about it to the actual attendance. In
between these two events Austin tries to explain the release pattern. publicity

and marketing efforts and the process audiences use deciding whether or not

to attend the movie.
Current movie audiences. argue Austin. are quite selective in deciding
which movie to see compared with the audiences of. say. the 1940s and
19503. To explain how this choosy audience selects what movie to see. Austin
proposes a model that involves the first three stages of the diffusion theory.
namely: (1) awareness and knowledge: (2) persuasion: and (3) decision. At the
awareness and knowledge stage. audiences “hear or read about” a particular
movie and perhaps some of its attributes such as its stars. the story or the
director. And the individual seeks more information concerning the innovation
( :— the movie) in order to proceed to the next step: Persuasion. At the
persuasion stage the consumer tries to form a judgment about the innovation
( § the movie) based on the things he or she has learned about it in the previous
a tage. The consumer. for example, may ask: “Do I really want to see that
movie with De Niro?” If his or her own response to that question is positive.

the process moves the individual to the third step: Decision. Decision simply

64

involves the acceptance or rejection of the innovation (ie. going or not going to
the movie). However. argues Austin. the decision to reject the innovation (going
to the movie) may change later on as the individual learns more about the
innovation and decides to adopt it (=attend the movie) say. later on. Austin
treats the remaining two steps of the theory. “implementation” and
“confirmation.” quite casually which only occur if and when the adoption has
been chosen. Consequently. he says. implementation is the preparation to
attend the movie and confirmation is the consumer's attempt to “reinforce the

adoption decision and to avoid or reduce any dissonance that might arise.”

For Austin (1991). the decision-making process in selecting which movie
to see is analogous to a camera lens that possesses a zoom function and a
series of focusing rings. The zoom motion accounts for the manner in which
people decide to go to a movie and the focus rings represent the array of
“conceptually similar attributes of or about a movie.” At the wide-angle
position. the audience member is faced with all the movie choices available and
the close-up setting decreases the number of options (See figure 3.1). Here
Austin identifies two types of movie attendance. First is the type that first
decides on the movie and then decides when to attend. For those. argues
Austin. the zoom moves quickly to the close-up end. Second. for those who
select going to a movie as an activity without a clear disposition to attend a
particular movie. the process begins with the wide-angle position and continues
with many or even all focusing rings and the decision depends on how well the
~E{.Q-eus rings line up. -

65

Figure 2.1

A PROPOSITIONAL MODEL OF THE MOVIE SELECTION PROCESS

 

 

 

   

zoom Lem -? NARROW
A ANGLE
WIDE ”‘0”
ANGLE “mm 3 U, r3 :0 MOVIE
8 § 3 § SELECTED
3 ‘< (b o
a 2 a: 2'- : a =-
3 9'. .<.. S '8 "’ 8
3. § ‘° g
MOVIE g: S 8
CHOICES «2 9' o

690900

\

I
688 90“

awalefiess /

figmedoe

 

 

 

Source: Bruce A. Austin. Immediate Seating: A Look At Movie Audiences. 1991

As the individual tries to narrow down the options (from wide-angle to
the close-up end). he or she proceeds through five elements. namely: (1)
publicity and advertising (trailers. awards. sequels. ads in the media): (2)
reviews (media reviews. MPAA ratings): (3) personal influence (word of mouth):
(4) story type (plot. genres): and (5) production elements (director. producer.
stars). Each element (focus ring) possesses a set Of variables and the
complexity of the element increases as the number of variables increases. The
production elements ring. for example. is the most complex. As seen in the
figure above. the focus rings are arranged in order of when audiences would
become aware of each and they correspond to the first three steps of the

diffusion theory: awareness and knowledge. persuasion. and decision.

66
While the “zoom analog” makes a reasonably strong model to explain the

individual movie selection process. the forced entry of a segment of the
diffusion theory. the innovation-decision process. makes it less powerful. The
major drawback of this forced entry of the diffusion theory to the “zoom model”
is the omission of the “adopter categories” and the time factor associated with
the diffusion analysis. The adopter categories represent many situational
factors that affect the adoptation of an innovation. For example. after
reviewing thousands of diffusion studies. Rogers (1983) concluded that
adaptation versus nonadoptation is related to (l) socioeconomic status. (2)
personality variables. and (3) communication behavior. These individual
differences. according to Howard (197 7). have a greater impact on non-durable
products than durables because “for durables there is somewhat closer
conformity. possibly because durables are typically more involving and more
likely to be ‘new’ than nondurables.”

In fact. Austin’s “zoom analog!” can stand alone as a model without the
diffusion theory. The attempt to enforce portions of the diffusion theory only
tends to weaken his model. As we will see in the next section. the crucial
variables in the Austin model are the same as the set of variables the
economic/business approach considered in their predictive models. thereby

suggesting a convergence or synthesis.

The Economic / Business Approach

The economic/ business approach examines the institutional factors
affecting the financial success of motion pictures. As noted earlier. the
primary concern of this approach is on decision making for a particular movie

but not why people choose to go to movies as an entertainment option. By

67

focusing on aggregate data and incorporating the supply side of the industry.
the economic/ business approach attempts to explore the variables that
influence the collective movie attendance decisions. the precursor to financial
success. While there have been several attempts to explain movie attendance
on economic terms over the past two decades (Garrison. 1971: Simonet 1977.
1987: Kindem. 1982: Dominick. 1983. 1987). Litman's collective research on
motion picture economics best illustrates the business/economic approach
(Litman. 1979. 1982. 1983: Litman and Kohl. 1991). Common to most of the
economic approach research is the use of rental revenues as the criterion (the
dependent variable) for movie performance and the use of quantitative

statistical techniques to test hypotheses.

Litman proposes an economic model to empirically test the theoretical
assumptions and industry folklore that underline movie attendance. While
accepting the importance of demand and the interaction between demand and
supply. Litman approach is more involved with the supply side elements of the
industry. Litman (1983: Litman and Kohl. 1991) asserts that there are three
crucial decision making areas that determine the financial success of a movie:
(1) the creative sphere: (2) the scheduling and release pattern: and (3) the
marketing effort. An incorrect decision in any of these areas can adversely
affect the financial success of a movie. The Litman model measures a set of
variables that are expected to affect each of these spheres using aggregate

industry data.

The creative sphere refers to the story type (genres: eg. drama. science
fiction. action-adventure. comedy. etc.. and pre-sold plots). Motion Picture

Association of America (MPAA) ratings (eg.. G. PG. PG-l3. R. NC-l7). director

68

and stars . production values (eg.. special effects. exotic locations). According
to Litman. stories that are familiar through other media have a natural
advantage of pre-selling the film and some genres appeal to a larger audience
than others. Reputable directors are not only able to put all the creative talent
together. but also have the necessary skill to bring outside financing to begin
the product. Although star worship changes from time to time and movies with
mega stars may flop. popular stars appear to have the power to attract
audiences to the movie theaters nationally and internationally. Big budgets are
capable of bringing higher production values to movies. which adds more
special effects. exotic locations and skilled technical talent to a movie. The
ratings assigned by the MPAA arguably send a message to the potential movie
audiences about the general applicability of the movie. Certain ratings codes.
such as PG (parental guidance suggested) argues Litman. are more sought after

by the producers as they seem to have the largest audience applicability.

The second area of importance is the scheduling and release pattern that
includes the influence of the distributor. release date and pattern. and theater
bookings. It is argued that major distributors with their “deep pockets” could
provide the necessary financial backing. and use their clout in the industry to
find preferential access to the exhibitors. Although there can be “sleeper hits”
in any given time of the year. a movie released in peak audience attendance
times of the year (ie. during summer months. around Christmas and Easter

holidays) can benefit from this natural attendance upswing.

Alternatives to the peak release are the exclusive and multiple bookings.
By first releasing the movie to a limited number of theaters in selected big

cities. the distributor can try to create a somewhat artificial demand for a

69

movie that presumably follows a positive word-of-mouth publicity. Multiple

releases allow the distributor to create a “temporary excitement” before

negative reviews and adverse word-Of-mouth penetrate the audience.

The final area of importance is the marketing effort. A movie. says
Litman. like any new product. needs to be marketed to the specific
demographic segments through all the mass media. The extent of the media
campaign is dependent upon the release pattern. budget. and the stars
involved and is aimed to entice the eager moviegoer to the theater. The
objective of the promotional campaign is to ignite the word-of-mouth which

would eventually replace advertising as the leading promotional agent.

Although not under the direct control of the distributor. the promotional
effort is also designed to benefit from critics’ reviews and being
nominated] awarded major awards like the Oscar. It is argued that positive
reviews and major award nominations/ receipts trigger the word-of-mouth

which in turn boosts the box-office.

In order to estimate the relative importance of each of these three
Spheres and the variables associated with them. Litman‘s model uses aggegate
industry data and estimates a revenue equation. Using regession analysis. it
then tries to explain the variance in rental revenues associated with the
Variables in the model. In its latest empirical testing the Litman model was

able to explain nearly one half of the variance in rental revenues (Litman and

Kohl. 1989).

70
Towards A Hybrid Approach

It is evident from the preceding discussion that both communication and
economic approaches help explain the substantially complex social
phenomenon of movie attendance. While each approach has its own merits.
neither one stands as a complete model that fully explains movie attendance
behavior. This is largely due to the preferences given to the data sources.
types of hypotheses being tested. inclusion and exclusion of certain predictor

variables. and the research methods used for data collection and analyses.

Specifically. there are several major reasons that justify the need of a
hybrid model in place of the communication and economic approaches. First.
as the survey of literature pertinent to the communication approach revealed.
the communication model does not take into account the structural changes in
the industry and the constantly changing audience demogaphics. It is evident
that an unprecedented number of people are now watching movies than any
other time of movie industry’s one hundred year history. Nevertheless. the
majority of viewing is taking place outside the traditional theater. For example.
rental of movie videos now surpass the theater attendance by a factor Of four
to one and this disparities continues to increase. Looking at the past few
years' industry performances tells that this is not merely an aberration but
rather a trend. These changes in the demand side have drastically altered the
decision making process of the supply side of the industry which is evident by
the way movies are sequenced and given time clearance for each successive
exhibition window. The new technologies. as Litman (1991) has noted. that
have been “blowing an ill wind for the motion picture industry ever since the

arrival of television in the late 19403.” brought a whole new generation of

71
competitors to the exhibition side of the business in the 19803. The new

competitors include home videocassette. premium pay cable. and pay-per-view
cable. Financially. home video has become the most important income source
for the movie industry. Rental revenues now account for only a third of the
domestic rental revenues. In this context. it is quite inadequate for an analysis
to not incorporate these changes in the industry as one finds in the

communication approach does.

Second. the communication approach also fails to accommodate the
demogaphic and lifestyle changes in the society. There is little doubt that
movies appeal to some demogaphic segnents of the society more than the
others. The rating system alone is a classic example of that. Besides the
appeal of specific movies to specific segnents of the audiences. demogaphics
appear to have an impact on where and when the movies are seen. For
example. it has been noted that as the baby-boom generation slides into middle
age. Americans will seek more of their entertainment at home. According to
the Consumer Expenditure Survey. reported in American Demographics. the
home entertainment trend is real-but not at the expense of public
entertainment. Spending on fees and admissions. for example. fell only slightly
between 1986 and 1991. while spending on other entertainment. much of
which is home-based. increased (American Demographics Desk Reference. July
1993). Moreover. householders aged 35 to 44 who are most likely to have
school-aged children spend the most on videotapes as well as taking their
children to the movies. This trend is evident from the recent success of G.
PG. and PG-13 movies and consequent response from the movie industry by

its willingness to produce movies that appeal to that demogaphic segment.

72
Movies with G and PG are also the biggest hits in the sell-through video

market. The communication approach overlooks the importance of these

institutional foundations and consequently becomes acontextual.

Third. although the communication approach is audience oriented. the
model has not yet been tested empirically in its entirety. As noted elsewhere.
the model is supported by evidence gathered from different studies and some
of those data sources date back to as late as 1950s. disregarding the drastic
changes taken place in the industry and the society over the past two decades.
Undoubtedly. the model accommodates some of the very important variables
that are crucial to explaining movie attendance. Therefore. it would be more
appropriate if they were tested together as a complete model to thereby

examine the validity of those factors in the present context.

The economic/business approach. on the other hand. has its own
shortcomings that can be overcome by converging its strong elements with the
communication approach and by employing a different research method. First.
economic models that rely on institutional data suffer from the proprietary
nature of financial information that is almost impossible to acquire or available
data that are incomplete. As Dominick (1983) has noted. “finding reliable
economic data on the motion picture industry is difficult and the information
that does exist is incomplete and in some cases contradictory.” For example.
only the movies that make at least one million dollars at the domestic box office
are available to the researchers. thus preventing the inclusion of movies that
are true financial “failures.” Furthermore. the dearth of revenue data on home

video industry force economic researchers to wait another few years until

73
reliable and complete data sources become available to include this important

segment in their research.

Second. although the economic approach promptly recognizes the
importance of structural changes in the industry. it was unable to
accommodate those variables into the model first due to the lack of reliable
data sources. and second the economic analysis looks to the supply side of the
industry. For example. as Litman and Kohl (1989) have noted. they were
unable to include advertising and promotional factors into their model due to
methodological reasons. Furthermore. some of the variables. such as the
impact of word-of-mouth advertising. demographic and psychographic factors
can only be measured from the demand side using the “self-reports” of the
audiences. As such. only a hybrid model would allow all advertising/ publicity
related factors to be included in a multivariate model to explain movie
attendance. Moreover. an additional advantage of the proposed hybrid model
would be the opportunity it provides to test the strength of the economic

factors using household data.

Despite the shortcoming discussed above. the communication and
economic models provide the only systematic approaches available to explain
the movie attendance behavior. Accordingly. this study recognizes the
importance of both. the communication and the economic/business
approaches. and attempts to advance the explanatory power of the two
research approaches by combining them and expanding them to include the
structural and demogaphic factors and the gowing home video industry.
Consequently. it appears that a convergence of two approaches would provide

a hybrid model that would benefit from the accomplishments attained by both

74

approaches. Furthermore. it is also evident that as the movie industry gow
into a multi-facet industry with several alternative windows. no single approach
or research method would be appropriate to single-handedly understand the
nature of the industry and the changing audience needs. Hence a need for a

hybrid approach.

The following chapter will elaborate on the research questions and

methods.

Chapter 3

Research Questions And Methods

This chapter will elaborate on the research questions. variable selection

and measurement criteria. the sample, and the data collection procedures.
Research Questions

Research Question 1: What factors have the most influence when
consumers are deciding which movie to see?

Rationale: As evident from the research cited in the preceding chapters, both.
communications theory approach and economic/ business approach. attempt to
answer this research question with different data sources and operational
procedures. The former approach focuses on individual decisions to choose
which movie to see but largely overlook the structural and institutional factors
affecting the decision making process. An additional weakness of the
communications theory approach is that it has so far been unable to empirically
test the explanatory or predictive power of the proposed model. The latter
approach focuses on the collective theater attendance of the moviegoing public
and uses rental revenues as the criterion for attendance. This approach uses
aggregate industry data and recognizes the importance of structural and
institutional variables on attending movies but fails to include them due to the

inadequacy of data sources and measurement difficulties.
75

76

The objective of this research question therefore. is to study the factors
affecting movie attendance decisions. from the demand side using the same
variables suggested by the economic and communication research approaches.

Research Question 2: What are the effects of the gowing home video
environment on movies seen in general and theater attendance in

particular?
Rationale: As evidence cited in chapter one has revealed. home video has
become the most important exhibition window for the movies. Rental statistics
indicate that more people watch movies on video than in the theaters or all
other windows combined. Domestically. it is the largest revenue center for the
movie industry. Yet. our knowledge about home video more or less ends at the
rental figures and revenue estimates. Although the broad financial impact is
evident. how the audiences perceive home video is largely unknown. Also not
known are the specific implications home video has and will have on theater
attendance. Although the economic research had constantly reminded us the
importance of gowing new windows on theatrical revenues. specifically the
home video. it was unable to measure their collective impact--a void this

research question tries to fill.

As the two research questions indicate. this study is theoretical and
exploratory. It is theoretical because it attempts to converge two theoretical
approaches in favor of a unified model to explain movie attendance and test it
empirically. As such. the first research question attempts to address the
theoretical part. It is exploratory. because it attempts to explore the specific
impact the home video industry has on movie attendance in general and theater
attendance in particular. It further explores the possibility of using the same

set of predictor variables that are being used to explain theater attendance on

77

video rental. As such. the second research question addresses the exploratory

aspects of the study.

Selection of Variables

As the two research questions deal with theater attendance and video
rental. two dependent variables will be used in the present study. The first
dependent variable (labeled as MOVIE#) estimates the average monthly movie
theater attendance of the respondents. Question number 1 in the survey
instrument is used to measure this variable (see Appendix I for the

questionnaire).

This variable will be used in two different forms in two types of
statistical analysis. First. as a continuous variable in multiple regession
equations and then as a discrete variable in discriminant analysis. The use of
this variable as a criterion variable in multiple regession required assignment
of approximate values to the variable by the researcher. As seen in the
questionnaire. if a respondent attends one or more movies per month. the
actual number is recorded and used. However. if the respondent's attendance
frequency is less than once a month. he or she is then classified into one of the
appropriate categories (i.e. “Once about every two to six months.” or “About
once a year”). To approximate a value that would be isomorphic to the real
values of these two categories. .5 and .l were assigned to the first and second
categories respectively. Naturally. those who “never” go to movies received a
true 0. Part of this measurement procedure is adopted from MPAA survey
instruments. Given the sporadic nature of theater attendance among the older
public. this seems to be the closest one can get to measure attendance on a

continuum.

78

The discrete form of this variable required no researcher assigned
values. In accordance with the industry definitions. those who go to the
movies at least once a month were categorized as “frequent” moviegoers and
those who attend less than a once a month as “infrequent” moviegoers. This
form of the variable effectively divided the sample into two groups and is used

as the discriminatory factor variable for discriminatory models.

The second dependent measure estimates the number of movie videos
(or laser discs) rented during an average month and is labeled as VIDRENT.
Question number 19 is used to Operationalize this variable. This variable will
be used as the criterion factor to estimate regession equations for video rental

patterns.

Twenty-two independent variables will be used in the present study as
predictors to explain theater attendance and video rental patterns. Twelve of
them are related to creative and promotional aspects of movies as defined by
Litman and Kohl (1989). The remaining ten variables measure demogaphics
and price factors. As the research questions of this study emerged from
economic and communications research and industry analyses. an effort is
made to accommodate all the possible variables suggested by the appropriate
theories and industry findings. Unfortunately there is no way to measure
certain variables in the Litman model from the demand side using attitudinal
household data. For example. the variables pertinent to the scheduling and
release pattern. such as the importance of release pattern. number of screens.
or the impact of production budgets. are impossible to Operationalize
meaningquy using household data without drastically changing the research

design. As such. the set of variables that come under “scheduling” sphere in

79

the Litman model were not attempted to be measured in the present study.
However. some new variables. derived from the analysis of home video
industry. are included to fully explore the home video audiences and its impact

on general movie attendance.

1n the following section. the independent variables will be further
elaborated with reference to the specific questions in the survey instrument

that are used to estimate them.

Creative and Promotional Variables

The importance of Stars. Director. Story Type. MPAA Ratings. Critical
Reviews. Theater Previews. Previously Known Story. Word-of-Mouth
advertising. wining or Nominations of Awards such as Oscar or Golden Globe.
Advertising in the Media. Star Appearance on TV Talk Shows. and the movie
being a Sequel. are the variables that will be used as creative and promotional
predictor variables of movie attendance and video rental. Most of these

variables have been used before. and each will be discussed separately.

Importance of each of these variables to the audiences is measured on
the basis of self reports. As question 5 of the survey instrument shows.
respondents were asked to rate how important each of these factors is to them
when they decide which movie to see. The rating scale ranged from O to 10
where O = “Not Important At All.” and 10 = “Very Important.” Variables are

listed in the order in which they appear in the survey instrument.

The first variable from the creative sphere is STARS. This variable is
expected to estimate the importance of a star or stars on the attendance

decisions. While the conventional wisdom always tends to attribute a movie’s

80

success to the star power. the actual contribution of “superstars” has changed
from time to time. For example. Litman (1983) found that the presence of a
star had no significant impact on rental revenues but Litman and Kohl (1989)
found that the star actors and top directors were significantly related to
financial success. Austin (1989) also asserts the significance of popular stars
on attendance but argues that stars alone are rarely an important factor in
attendance decisions. The ever increasing market value of “superstars” also
indicates the faith the industry has on the stars. The stars. therefore. are

expected to have a significant impact on attendance decisions.

Both. economic and communications research. recognizes the
importance of reputable director (DIRECTOR) on financial success and
attendance decisions. Economic research goes beyond the simple reputation of
the director to his or her ability to blend all the creative aspects of a movie
including the director’s reputation in industry circles to obtain outside financing
for a film project (Litman. 1989). While it is difficult to measure such
attributes of a director from the audience‘s perspective. only the important of

the director to the respondent on the attendance decision is measured.

In their last testing of the Litman model. Litman and Kohl (1989) used
fifteen categories to describe the story type or genres (STORYTYP) e.g.. Drama.
Comedy, Action Adventure etc. The economic research measures the impact of
each different genre category on rental revenues. However. testing the
importance of each specific genre category also creates some problems since
there are no established categories (Austin. 1991). As such. it is difficult to
develop reliable measures to assess the importance of each genre type to the

audiences as they often having difficulty identifying the proper genre except for

81

some overwhelmingly obvious categories such as Comedy. Western or Science
Fiction. This research. then. measures the perceived importance of genres on

attendance in general. not a specific genre.

The MPAA rating (RATINGS) - G. PG. PG-l3. R. NC-l7 - generally is an
indicator of the degee of sexual content. violence and harsh language in a
movie. Past research and industry preferences suggest the largest audience
appeal for PG and PG13 movies while the R. X and “not-rated” movies scare off
the audiences (Litman. 1983. 1989). Similar to the genres variable. economic
research also measures the importance of individual rating types on revenues.
Again. this research measures the importance of the overall concept of ratings

and not the likes or dislikes of specific rating types.

The recent success of movies that were adaptations of. or based on. a
previously known story (PRESTORY) that is familiar to the public shows the
positive impact the familiarity of the story line has on attendance. As Litman
and Kohl (1989) noted. the movies based on known stories (e.g.. successful
plays/novels. TV shows. cartoons etc..) have a built-in advantage over those
that start from scratch. The recent successes of The Fugitive (TV Show)
Schindler’s List (Book and the subject-Holocaust). suggest the importance of the
known story in attracting the audiences. They also give a head start in
marketing and promotions as audiences already possess some knowledge

about the story line.

The last variable associated with the creative aspect is the sequel
(SEQUEL). The attractive power of a movie that is a sequel of a one that the
respondent has seen and liked is measured. While the economic research put

sequel under the same category of a “previously known story/idea." sequel is

82

measured in this study as a separate variable with the potential or collapsing to

a scale if positive correlation exists between the two.

Litman and Kohl (1989) regard advertising and publicity as the key to
luring the avid movigoer into the theaters to begin the important word-Of-mouth
campaigr. However. they were not able to include either one of these variables
(advertising or word-of-mouth) due to the practical problems associated with
measuring them. The inclusion of this variable may then provide some
insights to their effects on attendance decisions. Two separate variables are
measured. one for advertising in the media (ADVTSNG) and the other for the

importance of other people‘s recommendations (WORDOFM).

Theater previews (PREVIEWS) have also not been included in economic
research. Austin (1991) considers theater previews or “trailers.” as a form of
advertising that address an already interested group. and he includes them in
his “zoom model” under the broad category of “Advertising and Publicity.” The

importance of this variable to attendance decisions is estimated here.

Winning or having been nominated for major awards. such as Oscar or
Golden Globe (AWARDS). is considered by Litman and Kohl (1989) to be a key
advertising/ publicity tool. The financial significance of this variable is well
recognized even in the trade circles. For example. the Advertising Age recently
observed that “Oscar nominations and eventual Academy Award winucan as
much as double a film's box-office business and provides strong mainstream
appeal for “art-house” pictures (Advertising Age Fax. February 9. 1994).

Litman and Kohl (1989) only included the nomination or wining of an Oscar in
Best Picture. Best Actor and Best Actress category and found the nomination to

one of these categories is significantly related to rental revenues. This

83

research attempts to take the awards factor one step forward and includes not
only Oscars but also other recognitions such as Golden Globe. as these awards

also receive increased media attention and subsequent publicity.

Appearances of stars on television show such as. Today Show. The
Tonight Show. or Late Night With David Letterman (TVPROMOS). is a new
variable included to assess the publicity value of star appearances on TV talk
shows on attendance. In recent years this has become such a crucial publicity
event that some studios book their big stars in more than one show in the
same week to coincide with the release dates. This is certainly not as
important as Awards on the attendance decisions but it would be interesting to

see whether this factor is at least registering in the audiences‘ radar.

Demographic, Media and Economic Variables

Seven demogaphic variables were included in the survey instrument to
be used as predictor variables for movie attendance and video rental patterns.
They are also expected to be used as gouping variables for bivariate analysis.
Age (AGE). level of education (EDULEVEL). and income (INCOME) were
measured using the same categories used by the annual MPAA surveys and
other standard measurement procedures used in mass media research. Three
binary (dummy) variables were used for marital status (MARITAL), respondent’s
gender (GENDER). and absence or presence of children under 16 in the
household (CHILDREN). The total number in the household (HHHEADS) is
included as a continuous variable. The last seven questions (32 - 38) in the

survey instrument are related to the measurement of demogaphic factors.

In addition to the demogaphics. several price and media related

variables were also included as possible predictor and grouping variables. They

84
included: subscription to cable television (CATVSUB - Q#1 l). respondent’s

willingness to attend more movies if the admission prices were lower (ADMSNS
- Q#30). number of movie channels subscribed to (PREMIUM# - Q#13). the
total cost for a movie outing (MOVICOST - Q#29). and an estimate of the
percentage of movies seen in the theaters out of all movies seen during the
past six months: (PERCENTG - Q#26). The inclusion of some of the media
penetration/ usage variables such as the number of movies watched on
premium channels. pay-per-view availability. usage and frequency of usage

(items 14 through 17) were primarily for descriptive purposes.

Additional questions were included primarily to compare video rental
behavior with theater attendance and to measure audience perceptions about
the home video window. Questions 6 through 8 and item 23 are designed to
this effect. Four Open-ended questions. adopted from Austin's (1989) research.
were used to measure what attracted the viewers to their last movie seen in a
theater and the last movie seen on video (items 2. 3. 21 and 22). Additional
open-ended items were included to explore what the audiences most and least

like about theater attendance and video rental (items 9. 10. 27 and 28)

The expected relationships between independent and dependent
variables are largely governed by (a) the theoretical suggestions: (b) the current
industry structure: and (c) the specific window (movie or video) the variable is
used to explain. As such. the expected relationships will be discussed with the

results in the following chapter.

85
Sample and Data Collection

The sample for this study was drawn from the general public15 to
represent a cross section of the movie going population. Although it is more
painstaking than a readily accessible student sample. pure reliance on a
convenient student sample might force the research to compromise the validity
of the findings due to the atypical nature of such a sample. Moreover. it is
believed that a randomly drawn representative sample will allow the use of
inferential statistics meaningfully without violating the stipulated assumptions
of the multivariate tests. Consequently. a random sample of general population
appeared more appropriate to obtain more generalizable estimates of the

variables.

Telephone survey method was employed to collect data because it
enables the researcher to reach the respondents in a relatively economical and
timely way still assuring a relatively higher data quality and an acceptable
completion rate. Two telephone interviewers with previous experience in
telephone interviewing were employed to assure the data quality”.
Interviewing sessions were stretched to a 7 day period to allow time for
maximum number of call-backs. again. to assure the data quality by increasing
the response rate. While most interviewing took place between 6.00 p.m. and
9.30 p.m.. on weekends interviewing began at noon and ended at 6.00 p.m.
On two week days one interviewer worked from 9.00 a.m. to 5.00 p.m. to
reach those people not available during the evenings. Interviewers were

directed to call each number a minimum of four times.

 

15Residents of East Lansing Local Access Transport Area (LATA) that includes: East
Lansing. Lansing. Okemos. Holt. Mason. and Grand Ledge.

16Both interviewers were graduate students in the College of Communication Arts and
Sciences who holds part time employment in research institutes doing survey research.

86
Random digit dialing. using the local LATA prefixes. was used to select

and reach the sample. While random digit dialing is one of the most
comprehensive method for selecting a random sample of telephone numbers. a
large number of the numbers generated by this method tend to be invalid
because some phones have been disconnected. some numbers generated have
not yet been assigned. and for other reasons such as. the numbers assigned
for businesses and commercial establishments. The adverse effects of this

drawback are. however. not on data quality but on time and money.

Although most of the questions were derived from previous studies and
information gathered from two focus goup sessions conducted in December
1993. two pilot studies were conducted to refine the survey instrument on

small samples (~20 subjects each) from the target population.

Telephone Survey Outcome

Table 3.1 presents the outcome of the telephone survey results and the
completion rates. A total of 845 telephone numbers were called with 490
successful contacts. Of the 490 eligible respondents that were contacted. 414
(84.5%) ageed to take part in the survey while 76 (15.5%) refused. However.
48 of the 414 respondents who ageed to take part in the survey said they
haven’t seen a movie either in a theater or rented one on video during the past
six months. As such. their interviews were not continued. so the effective
sample size was 366 (414-48). As expected from a sample of random digit
dialing, a large amount of numbers (181) were either not yet assigned.
disconnected or assigned to business or government establishments. The 58.7

percent response rate appears to be in the higher end for a telephone survey

87
and might be attributed to the higher concentration of college students in some

of the area prefixes in the sample.

The results of the study will be presented in the following chapter.

 

 

 

 

TABLE 3.1

TELEPHONE SURVEY RESULTS

FREQUENCY PERCENTAGE

COMPLETIONS 366 58.7
REFUSALS 76 12.2
NO ANSWER1 181 29.1
INELIOIELES2 174 —
EXCUSIONS3 48 —
TOTAL 845

 

1No Answer, Amwering Machines, Call Backs, Busy Signal, Language barrier
2Dr'sconnects, Not in service, Business

3Respondents who hadn't seen a movie in a theater or on video in previous 6 months

Chapter 4

Results

In this chapter, the results of the study are presented in the following
order. The first section will present the sample demogaphics followed by a
survey of the movie attendance patterns presented with the aid of univariate
and bivariate analyses. The third section will be similar to the second but the
focus will be on home video window. The results of multivariate statistical

analysis are presented in the final section.

Sample Demographics

Table 4.1 presents the summary statistics of major demogaphic
variables. Population distribution for major demogaphic variables are also
presented in Table 4.1 based on 1992 census data. Of those interviewed. 56
percent were females and 44 percent were males. The sample seems to
comprise about four percent more females than the actual population it expects

to represent.

About one-fifth of the sample was 16 to 20 years old and the age
gouping of 16 to 29 year olds accounted for slightly over one half (51.4%) of
the sample. The age categories used in this study were adopted from the
Motion Picture Association of America’s (MPAA) annual motion picture

attendance survey.

89

 

 

 

 

 

 

 

 

 

TABLE 4.1 SAMPLE DEMOGRAPHICS
VARIABLE PERCENT FREQUENCY POPULATION
(N-366) DISTRIBUTION'
GENDER
MALE 44% 160 48%
FEMALE 56% 206 52%
MARTIAL STATUS
SINGLE 59% 216 51 %
MARRIED/ LIVING TOGETHER 41% 150 49%
CHILDREN UNDER 16 IN HOUSEHOLD
YES 28.7% 105
NO 713% 261
AGE
1620 205% 75 15-19 11.2%
21-24 22.4% 82 20-24 14.4%
25-29 085% 31 25-34 221%
30-39 18.6% 68 35-44 19.7%
40-49 16.1 % 59 45-54 12.3 %
50-59 05.7% 21 55-64 8.8%
60 OR OVER 08.2% 30 65 > 11.5%
W
UNDER $15,000 303% 111 UNDER $15,000 21.3%
515,001-525,000 11.7% 43 515,001-525,000 16.7%
525,001-535,000 12.6% 46 525,001-,535000 16.2%
535,001-545,000 10.7% 39 535,001-550,000 19.8%
545,-.0015550m 09.3% 34 $50,001-ABOVE 26.0%
555,-,001565000 05.7% 21
$65,001 AND ABOVE 12.0% 44
REFUSED 07.7% 28
EDUCATION
LESS THAN HIGH SCHOOL 01.1% 4 15.7%
HIGH SCHOOL COMPLETED 10.7% 39 27.6%
SOME COLLEGE 51.9% 190 24.1 %
COLLEGE GRADUATE 21.6% 79 145%
POST GRADUATE 14.8% 54 10.2%
ASSOCIATE DEGREE _ -- 7.9%
MEAN MEDIAN STD DEV
NUMBER OF CHILDREN 1.9 2.0 .99 'BASED ON 1992
NUMBER IN HOUSEHOLD 2.89 2.0 1.57 CENSUS DATA

 

WI

90
A majority of the households (54.6%) had annual incomes of $35,000 or

less and 7.7 percent of the respondents refused to reveal their incomes--the
only question in the entire survey where respondents refused to answer. Fifty-
nine percent of the respondents were single. and the remainder were either
married (38.5%) or living together (2.5%). More than one-fourth of the
households surveyed (28.7%) had children under 16 living with them. The
average number of children in households was 1.9 (SD=1.0). Overall. the
average household had 2.89 (SD=l.57) persons. slightly larger than the

national average of 2.5 persons.

As expected in predominantly a college community. there was a high
education skew in the sample. Nearly 90 percent of the sample had at least a
partial college education and almost 15 percent of the respondents had post-
gaduate training. Overall. the sample appears fairly representative of the

population it is drawn from.

An Overview of the Movie Attendance Pattems

The purpose this section is to survey the general moviegoing patterns of
the respondents. The findings reported in this and the following sections will
be complementary to the multivariate analyses in answering the two research

questions central to this study.

Using theater attendance frequency as the criterion. the movie industry
categorizes moviegoers in four broad goups: Requent (those who attend at
least one movie per month): Occasional (those who attend at least one movie in
two to six months); Infrequent (those who attend less than once in six months):
and Never (those who reported not going to the movies at all) (MPAA. 1990).

When the same criterion is employed. this study found the majority of the

91

respondents to be “frequent” moviegoers (54.3%). about 30 percent
“occasional.” and 8.2 percent “infrequent.” Only 7.4 percent of the
respondents reported that they “never” go to the theaters to see movies.
Nonetheless. those who never go to a theater to see a movie were kept in the
sample because subsequent probing found this goup uses other windows.
specifically home video. to see movies. The average adult frequents a theater
slightly more than once a month (Mean=l.3). A little more than one-half of
those surveyed (52.7%) attend movies one to four times a month. Perhaps the
high skew is due to the somewhat over-representation of the “younger” people

(16-24) in the sample.

The overall average attendance. however. appears to disguise the
specific movie attending habits. A closer look at the different demogaphic
subgoups shows that these goups significantly deviate from the average
attending habits. For example. those who are single frequent movies
significantly more often than their married (or living together) counterparts
(Group Means: l.51/.90: t=4.95: df=360: oc=.001: two-tailed test). As
expected. age makes a significant difference in movie attendance as well.
Those who are in the 16 to 29 year age category attend significantly more
movies than people 30 years or older (Group Means: 1.58/ .92: t=5.18:
df=350; oc:.001: two-tailed test). It is no surprise then that the movie
industry estimates show this age goup (16 to 29) accounts for nearly one-half
of the total domestic movie attendance (MPAA. 1992). Also as expected. those
who have children living with them go to the movies significantly less often
than those without children (Group Means: .99/ 1.37: t=2.89: df=244: °C=.01:

two-tailed test).

1.1."th

92
Further differences are also present with regard to the subscription of

premium cable channels. Those who subscribe to premium cable channels
(particularly movie channels such as HBO. TMC. or Cinemax) tend to frequent
movies siglificantly more than those who do not (Group Means: 1.62/ 1.14:
t=2.55: df=l l7: oc=.01: two-tailed test). Although the cable TV subscribers.
VCR/Laser Disc owners and males tend to go to the movies more often than
their respective counterparts (Group Means: 136/1.21: 1.35/ 1.21:

1.39 / 1.28. respectively), the differences are simply due to chance than to a
statistically meaningful regularity. These findings are consistent with annual
MPAA survey results where neither gender nor cable subscription or VCR

ownership had significant effects on theater attendance.

4. 3 Importance of Creative and Marketing Factors on Movie Attendance

As the research cited in preceding chapters pointed out. both
communications and economic research traditions indicate the importance of a
set of creative and marketing variables on individual and collective movie
attending decisions”. Table 4.2 presents the importance assigned to those
factors by the respondents. Variables are listed in order of their importance to
the entire sample with corresponding means and standard deviations.
Subsequent columns show the respective mean values for four demogaphic

sub goups and the results of t-tests to assess the group differences.

 

”The 12 variables listed in Table 4.2 are broadly divided into two goups adopting
Litman's (1989) and Austin's (1991) classifications as “creative” (STARS. DIRECTOR.
STORYTYP. PRESTORY. SEQUEL. REVIEWS. RATINGS) and “marketing” [WORDOFM.
PREVIEWS. ADVTSNG. AWARDS. and TVPROMOS)

Table 4.2

93

 

IMPORTANCE OF CREATIVE AND MARKETING FACTORS ON MOVIE ATTENDANCE

(SCALE: O'NOT IMPORTANT AT ALL; IO'VERY IMPORTANT)

 

 

 

TTLSAMPLE GENDER MARITAL CHILDREN AGE

FACTOR MEAN (SD) M/ F MRD/ SGL YES/ NO 16-29/ 30>

1. STORYTYP 8.39 (1.44) 818/854 857/826 857/832 810/869
7, WORDOFM 7.44 (1.71) 7.44 /7.44 7.75/7.23 7.77/7.31 7.22/7.67
3. STARS 7.05 (2.11) 7.07 / 7.02 6.97/ 7.10 6.98 / 7.08 7.15/ 6.93
4. PREVIEWS 6.57 (1.82) 6.42/ 6.68 6.42/ 6.68 631/643 6.73 / 6.39
5. REVIEWS 5.81 (2.31) S.61/5.98 6.27/5.50 834/860 821/846
6. ADVTSNG 5.73 (2.01) 5.93 / 5.58 5.65/ 5.80 5.89/ 5.68 5.90/ 5.55
7. SEQUEL 4.87 (2.40) 4.81 / 4.91 4.97/ 4.80 5.05/ 4.80 4.68/ 5.07
s. PRESTORY 4.79 (2.27) 4.70/486 807/460 4.79/4.79 840/820
9. AWARDS 4.41 (2.42) 456/430 471/420 452/437 418/466
10. TVPROMS 4.10 (2.25) 394/414 3.77/4.25 3.86/ 4.16 838/331
11. RATINGS 3.80 (2.96) 846/406 467/820 817/284 808/456
17. DIRECTOR 3.77 (2.53) 4.01 /3.58 3.65/ 3.85 353 / 3.87 885/868
Group Sizes 366 160/ 206 150/ 216 105/ 261 188/ 178‘

 

BOLDmr-s INDICATETHE MEAN DIFFERENCES THAT ARE SIGNIFICANT AT cc 2.05

ALL TWO-TAILED TBTS.

As seen in the Table above. story type (STORYTYP) topped the list and

the first six variables received above average ratings. Also evident is the

mixture of creative and marketing factors in the rankings. What is more

important. however. is the siglificant differences found between the sub

goups. For example. although it is in the lower end of the list. the importance

of MPAA ratings (RATINGS) significantly differs from goup to goup. As

evident from the goup means. older. married. females with children found

MPAA ratings significantly more important in their movie attending decisions

than their respective counterparts. Likewise. those who are married also

94
found STORYTYP. other people‘s recommendations (WORDOFM). reviews in

the media (REVIEWS). wining or nomination of awards (AWARDS). and the
movie being based on a previously known story (PRESTORY) as significantly

more important than those who were not married.

It is revealing to find evidence to the effect that the importance of
creative. and marketing variables are related to demogaphic and other life
style factors. Correspondingly. “the young.” “the single.” and “the ones without
children” are appear to place a lesser value on these factors than their
respective counterparts. The empirical evidence simply shows how

indiscriminant the young moviegoer is.

The Social Context of Movie Attendance

An important factor in movie attendance is the social context in which it
takes place. Only a very small minority (5%) of the respondents goes to the
movies alone. For the vast majority it is a social activity. Most young adults
(16 to 24) go to the movies with friends or a date. and for those who are 30
years or above. going to the movies is a family activity. The social appeal of
moviegoing goes beyond doing something with the family and friends. For
some. it also means being part of an “unknown” crowd. For example. 53
percent of the respondents cited “going out” or “being in the crowd” as the

most important reason for going to the theater to see a movie.

The social nature of going to the movies appears to cost the audiences
dearly. On the average. it cost them more than 12 dollars to go out to see a
movie (Mean $12.46: SD=$7.70). Again. the average tends to conceal the
details as different subgoups spend well above the sample average. Naturally.

the cost is a function of the magnitude of the social ties. As social ties gow.

95

so seems to be the cost. Those who are married. for example. spend twice as
much as those who are single for a movie outing ($24.19 vs. $12.16: t=4.42:
df=333: oc:.001: two-tailed test). Similarly. those who are 30 years or older
spend $21.50 for a movie outing compared to the $12.80 the “younger” spend
(t=4.47: df=287; oc=.001: two-tailed). As expected. larger differences also
exist between those who have children living with them and with those who do

not ($25.89 vs. $13.64: t=8.03: df=112: oc:.001: two-tailed test).

 

 

 

TABLE 4.3
LEAST LIKE ABOUT GOING TO A THEATER TO SEE A MOVIE
FREQUENCY PERCENT AGE

PRICE (ADMISSION ANDCONCESSION) 153 45.1%
NOISE (PEOPLE TALKING, RUDENESS) 64 18.9%
COMFORT (UNCOMFORTABLE CHAIRS, STICKY FLOORS) 35 10.3%
CROWDS 32 9.4%
LOGISTICS (FINDING A SIITER, PARKING) 18 5.3%
HAVING TOGOOUI‘ 14 4.1%
NONE 10 2.9%
OTHER (PREVIEWS, SHOW TIMES ETC.,) 13 3.9%
(bl-339)

The sigiificance of price factor also emerged elsewhere in the study as
well. When asked to name what the audiences least liked about going to a
theater to see a movie. the high ticket and concession prices topped the list as
the most cited reason (see Table 4.3). Furthermore. 75% percent of the
respondents said they would go to the movies “more often” if the ticket prices

were lower. As such. it seems reasonable to believe that it is mostly the high

96

costs and other “inconveniences” associated with going to the theater. not the
movie quality itself. that keeps the majority away from the theaters. It also
explains why the adults. as a goup. are more cautious when it comes to
theater attendance decisions. For those who are older the consequences are
more expensive and time consuming than for the younger audiences who have
more time. fewer social obligations and. above all. are likely to get cheaper

admission prices as well.

What then do the movies offer to outweigh the audiences' antipathy and
attract them to the theaters? The survey results appear to shed some light into
this area. Responses to the open ended question that asked to single out the
most important reason for going to a theater are categorized in Table 4.4. More
than 95 percent of the audiences ages on four major reasons: escape (43.8%):
technical superiority (31.7%); to see it now (10.3%): and the theater
atmosphere (9.1%). The escape represents such reasons as “need to go out.”
“to relax.” “to unwind.” “to get away from kids.” and “to get away from daily
routine.” Technical reasons include the “big screen.” “superior sound quality.”
and the “life like” experience only the theater can offer. About ten percent of
the audiences go to the theaters because they want to see the movies “now” or
see it “before the others see it.” About the same number of respondents like
the theater ”atmosphere” where they can be “among the crowds.” These
patrons. as economists would like to call it. perceive a degee of utility from the
theater going experience that offsets the inconveniences (efforts) and costs

(admission and other) to watch a movie in a theater.

97

 

 

 

 

TABLE 4.4
MOST IMPORTANT REASON FOR GOING TO A THEATER
FREQUENCY PERCENTAGE

ESCAPE (RELAX, UNWIND, GET AWAY) 145 43.8%
TECHNICAL (BIG SCREEN, SOUND QUALITY) 105 31.7%
TOSEEIT Now 34 10.3%
THEATER ATMOSPHERE (BEING INTI-IECROWD) 30 9.1%
NONE 7 2.1%
NOTONVIDEO 4 1.2%
OTHER 6 1.8%
(hi-331)

A rather qualitative measure. adopted from Austin‘s research. is used to
look into the reasons surrounding the selection of a specific movie. A large
majority of those who have seen a movie (81.2%) during the past six months
did remember the movie by name. Table 4.5 presents the reasons that
attracted audiences to the movie they have last seen”. While the list contains
more than ten reasons. most of the audiences apparently ages on a few
reasons. Among them. the stars. story line (the plot) and the story type
(genres) contributed to about 50 percent of the attendance while advertising
and publicity related factors accounted for about 25 percent of the attendance.
The Obvious limitation of this measure is its exclusive reliance upon the last
movie attended. which may or may not be a typical movie. For example. the
measure can be largely skewed if an “event like” movie. such as Jurassic Park.

dominated the theaters during the period measurements were taken.

 

13‘th Table aho presents the reasons for video rental

98

TABLE 4.5

 

FACTORS ATTRACTED TO THE LAST MOVIE / VIDEO

 

WHAT ATTRACTED TO THE LAST ......

VIDEO

THEATER

 

FREQUENCY PERCENTAGE FREQUENCY PERCENTAGE

 

STAR 65
ADS/ PROMOS / REVIEWS/ PREVIEWS 16
STORY LINE 33
SOMEONE TOOK/ WANTED TO SEE 11
STORY TYPE 18
KIDSWANTED TO SE 18
RECOMMENDATIONS 19
AWARDS / NOMINATIONS 0
KNEWTHESTORY s
SEQUEL/FX/PREQUEL 5
DIRECTOR 4
OTHER 4
SEE AGAIN 16
ATTRACTIVE COVER 4
MISSED AT THE THEATER 2
(bl-222)

29.3%
7.2%
14.9%
5.0%
4.9%
4.9%
5.2%
0%
2.3%
2.3%
1.8%
1.8%
7.2%
1.8%
.9%

60
50
47
27

8

UTO‘UICDQ

NA
NA
NA

(N'273)

22.0%
18.3%
17.2%
9.9%
9.2%
7.3%
5.1%
2.9%
2.2%
1.8%
1.8%
4.0%

NA
NA

 

The Other Windows: Watching Movies At Home

Cable television is universally available in the sampled universe and the

subscription rate stood at 83 percent. About 30 percent of the cable

subscribers also subscribe to at least one of the movie channels (HBO. TMC.

Cinemax. Showtime or Disney). and premium channel subscribers watch an

average of 6.7 movies on these channels per month. While a large majority of

the cable households (82.3%) do have access to Pay-Per-View (PPV) services.

requesting movies on PPV does not appear to be a popular activity. Only 14

percent of the PPV accessible households had ever requested and paid for a

99

movie on PPV and 80 percent of those who do. do so only about less than once

a month.

Conversely. home video emerged as the most popular window for the
movies. An overwhelming majority of the households has a VCR (86.1%) and a
small percentage of those households (3.6%) are also equipped with a Laser
Disc player. Interestingly. all the laser disc households do have a VCR as well.
The average monthly rental of 4.1 videos is more than three times as much as
the average theater attendance of 1.3 per month. The average audience for a

rented movie stands at 3.3 people per household.

A considerably large proportion of the video audience appears not to
overlap with the theater audience. When asked about the frequency of renting
a movie that was already seen in the theater. a large majority (84.2%) reported
that they “rarely” or “never” rent the movies already seen and 11.7 percent said
they only do so “about half the time.” To build upon this measure. a follow-up
question asked about how Often the respondents wait to see a movie on video
rather than going to the theater. Nearly one-half of the people who rent videos
do wait for the video rather than going to the theater “always” or “almost
always” and a third (33.9%) do so “about half the time.” In addition. the
majority of the moviegoers (56.2%) would rent the video instead of going to the
theater if a movie is simultaneously released on video with the theater release.
and 56 percent are willing to pay more for such an early release than what
they pay now to rent a movie. Even if these estimates are marginally '
acceptable. home video has undoubtedly become by far the largest window for
the movies not only by mere penetration and usage levels but also how people

perceive it as an independent exhibition window for the movies.

100

There are some noticeable similarities and differences between the
theater and the home video windows. As shown in Table 4.5. the reasons that
attract audiences to the theater are more or less the same for video as well.
One noticeable difference is the impact advertising and publicity strategies
have on the two windows. While advertising and promotional factors are the
second biggest attraction for theater attendance, only 7 percent of the video
rentals were affected by those factors (see Table 4.5 for a comparison). This
seems quite normal because. unlike the theater release. the video release
hardly has advertising budgets or promotional strategies built into it. 19
Moreover. the effects of reviews largely dry out by the time the video hits the
stores. This finding also justifies the recent attempts by the industry to
narrow the passage of time between the theater and the video release to
benefit from the rippling effects of advertising and other promotions initially

done for the theatrical release.

The most significant difference between the home video and theater
windows surfaced when the audiences were probed about the most important
reason for renting a movie. It is crystal clear why people rent movies: price
and convenience. As Table 4.6 shows. virtually all video renters (91.5%).
prefer to rent because watching at home is overwhelmingly convenient and far
less expensive. As the respondents put it. they can “watch it at the
convenience of home.” “can watch any time I want.” “watch it without leaving
home” (categorized as “convenience”) and “it‘s cheaper.” “don‘t have to pay high

ticket and concession prices” or simply “the price” (categorized as “price”). A

 

19Except for most of the Disney movies and “mega-hit” titles such as The Fugitive and
Mrs. Doubt/ire where the video release B primarily intended for the sell-through market
with low unit prices.

101
self estimate of the respondents finally revealed that only 36 percent of the

movies seen during the past six months took place in a theater - hence it is

clearly a less important movie window than it was 10 years ago.

 

 

 

TABLE 4.6
MOST IMPORTANT REASON FOR RENTING A MOVIE
FREQENCY PERCENTAGE
CONVENIENCE 181 59.3%
PRICE 98 32.1%
WATCH WHAT MISSED IN THEATER 11 3.6%
WATCH AGAIN 6 2.0%
OTHER QAMILY EVENT, BAD THEATERS) 9 3.0%
(bl-3(5)
Results QfMultivariate Analyses

This section attempts to supplement the findings presented up to this
point with more sophisticated multivariate analyses to assess the collective
impact of the independent variables to explain the variability associated with
theater attendance and video rental. Two separate sets of multivariate
equations will be estimated--one for the movie attendance and the other for the

video rental.

Table 4.7 present simple correlations for the entire set of independent

variables with the two dependent variables”. Theoretical justifications for

 

20These simple correlations were obtained without any treatment to the missing values.
As missing values will be substituted for regession analysis. the strengths and the
significance levels of the present relationships may change as missing value treatments
efl'ectively change the sample composition.

102
selection of these variables were discussed in the preceding chapter and the

direction of the relationship between independent and dependent variables will
be discussed separately as their expectations may vary from one dependent
measure to the other.

As seen in Table 4.7 only four of the twelve movie quality and marketing
variables (DIRECTOR. STORYTYP. REVIEWS. AWARDS) are siglificantly
correlated with theater attendance (MOVIE#). However. given the fact that this
survey was conducted in March. 1994 during the weeks that surrounded the
Oscar nominations and actual awards ceremony. it appears necessary to take
some precautionary action before using this data in subsequent analysis.
There is a possibility that the respondents‘ self reports could have been
clouded by the events occurring at the time of the data collection. Specifically.
media hype about the Oscars was at its peak and so were the highly praised
reviews in all the media for the two prominent Oscar attractions. Schindler's
List. and Philadelphia and director Steven Spielberg. As such. it was decided
to weigh the relevant data to discount the “Academy Awards factor” of three of
the four variables concerned here (DIRECTOR. REVIEWS, and AWARDS). An
arbitrary weight of negative15 percent was assigled for those cases who had
seen either one of these two movies”. However. although the magnitude of
the coefficients had shown some positive and negative effects, the weighted
variables did not significantly alter the results of either correlation statistics or
regession analyses. Consequently. it was concluded that there is not enough
evidence to believe that the performance of these variables was signifiéantly

affected by. what might be termed the “history effects.”

 

21Fortunately. this information was available and 31 respondents had seen Schindler's
List and 44 had seen Philadelphia.

103

558.563.1285 5 3. S 585.85 u 28.

 

 

beau. Neae. vane.
e__~. new.. 5.4—.
~4~_. a.m.. Rha_.
”was. N445. axes.
_d_o. -c_. smeo.
cdc_. ann_. 545°.
38. 28.- 82.-
have. _w~_. memo;
danc- omNe. hch.
was". wan.. «Nan.
mhvn. odes. nn5_.
me_~. 5.4". made.
I) ”.mu. c.4n.
-. 568..
mcs<3< 2 docs? 5291s.

 

«mm—r ammo. womor o. 8. mean. 38.- 28. :8.-
Ehor 55:. 33. San. mmoor mér =42. ammo.-
mnm— .- was” amen. mean. we .o. 38. ”mn— .- «nun.-
mwvo. gho. 33. Nwoor mmno. :8. «an. coco.
ammo. 2.2. Run. 3%.. on _o.- 2 nor sown. we _ .-
3 _ _.- 3:. 2:". mono. ans.- oaoor o: _. woos.-
m n no. m m2..- hwmor manor ~68.- wo. or RON— . owao.
2.8. was. new". ago.- Nmmor 28. one". 38.-
Nmoo. mover woaor «So. mas.- n _ 8.. N23.- been.
mono. veg. swan. Soc. 2 _ _. woe". manor :8.-
Ncmo. seen. :8. oz?- h—v—. :3. 3.3.- 28.
«gm. 3:. 34:. ”to. N. 8. ani. obmor sewn.
cm 5. 3K". an. _. owns. «New. 2.5. Soc. amen.
ne—N. mama. e: _. moon. sews. wee. wooo. 38.
~36. amen. ago. mnoo. who". 2 2. 38. mace.-
l nut. 32. gnu. 2 so. hoes. homo. coco.
I. noon. «mg. ovwo. ~52. 3.8. was .-
Il moi. 3.8. S _o. awmo. goo.-
l.- 3.3. 2 8.. ago.- owS .-
-..I 3:. cos. 9.3.
.1 :8. Sec.
I- came.
mBSmmm mBmSmm mOZ—hé ggg «CPO-5:0 macaw F2555 emu—>02

ZOE<UDQM

mEOOZ—
m0<

mo<mIII

meQdIU

sen—.252
«ZmEn?

HmOUSO—z
OthUmmm

SODOmm

mOZOmm>F

02mgfl<
QO<3 <

SEOQKOB
>mOHmmmm
m3m=>mmm

m3m:>mm
mozpé

m>k>mOPm

MOHUmEQ
mass-m
h2m¢n=>

.0 II -VO-‘

 

KE<2 ZOF<AmMMOU man—EB
h... waned.

104

.22 3.5 oz. .285 a n... a .5955; - 2cm

 

 

£2... 8.... m8...- ....:. p.m.". .3..- a2... .2..- 88.- 2.....- :3...- 2220......
I 8?: 2...”. $2.. 8.... $3.. 33. 3......- owefi 8......- ........- ”.282.
I .3... we... 3...... can? .22. 2.8.- .2... S...- on... mo...

I 23.. Sam. SS. Eh. 2.....- :2. :8. :3... 85:2:

| 2.3.. .8...- nnnn. 9.2.. 2...... .8..- 8.... 2.23.20

I E...- 83. 3.2.. .5... 2o...- 38- .5232

I N8... 3....- 3... 3...... 5.... .2929.

I .8". E...- .8...- m..... $850.2

I 2.....- .2...- Nnc... 828......

I 2.2. 3.... 250....

I 25. 829...:

I @2222

8.23...

2.09.03

205......

was...

was...

82.25.

.>h>2ohm

208.25

925

5.35

.282. - mo... mammmm 2.5-=20 .5222 95.9.. 28502 82.8.... .msmmm .2925... 2239.. I .-

 

..._.....oo Ev 032,—.

l 05
Three of the other independent variables. marital status. having children

in the household. and age. (MARITAL, CHILDREN. AGE) are significantly
negatively correlated with the theater attendance (MOVIE#). Their negative
relationship with the movie attendance is within the range of theoretical
expectations as an “elder.” married populace with children tend to attend
movies less frequently than their respective counterparts. Although not
statistically significant. the cost of going to the movies (MOVICOST). willingness
to go to the movies if admission prices were lower (ADMSNS). and the number
of people in the household (HHHEADS) are within the range of expected
theoretical directions. No a priori direction is expected for the variable
household income (INCOME). INCOME is negatively correlated with theater
attendance but shows a positive correlation with video rental. suggesting the
higher the income the lower the theater attendance but consequently. the
higher the video rental. This is a somewhat awkward relationship given the
fact that going to the theater to see a movie is the more expensive option

between the two.

AGE is also significantly negatively correlated with video rentals but the
relationship is weaker than it is with movie attendance. While there is no a
priori direction for this relationship. it is reasonable to expect a positive
relationship between AGE and VIDRENT as the price and time sensitive “elder”
population may find it more “convenient” and less expensive to rent than to go
to the theater. As expected. theater attendance percentage (PERCENTG)
shows a strong significantly negative correlation with VIDRENT. Although not
significant. unlike with theater attendance. marital status (MARITAL) is

positively correlated with VIDRENT suggesting a tendency among the married

106

to rely more on home video window than on theater for their movie viewing.

A salient factor in the correlations results is the very weak associations
between the twelve movie quality and marketing variables and video rental
patterns. Not only do they lack significance. their correlation coefficients are
considerably low in many instances reaching near zero correlations. On the
contrary. demographic and economic variables are more strongly related to

video rentals some relationships gaining statistical significance.

Data Reduction Eflorts

Given the merits of predictive models with fewer variables but greater
explanatory power. several efforts were made to reduce the number of
explanatory variables along the theoretical lines by making composite
measures. Existing inter-correlations among the variables also encouraged
exploration of the possibilities of composite measures. As a first step. a
Varimax factor analysis was performed using all the twelve creative and
marketing variables. The goal of this was to represent relationships among
sets of variables parsimoniously. That is. if variables related to. say, the
creative sphere of the movies such as director. stars. story type were related.
it is expected that factor analysis would allow explanation of the observed
correlations using as few factors as possible. consequently allowing estimation
of predictive models with fewer variables. The results of the factor analysis
are presented in Table 4.8 in Appendix 11. As seen in the Factor Matrix. the
results were not very promising in theoretical or statistical sense. First, while
there were four factor loadings. the factors were theoretically meaningless and

largely uninterpretable. For example. the six variables that loaded on Factor 1,

107
STARS. DIRECTOR. REVIEWS. PRESTORY. AWARDS. and ’I'VPROMOS. are

not theoretically related. Similar is Factor 4 where WORDOFM and SEQUEL
showed heavy loadings. Theoretically. a variable like word-of-mouth makes
more sense if loaded with Factor 2 where the advertising and publicity
variables show heavy concentration instead of Factor 4. Factor 2 makes some
theoretical sense as PREVIEWS and ADVTSNG show a strong correlation.
However. it had left other advertising/ publicity related variables such as
WORDOFM and TVPROMOS out. Moreover. six out of twelve variables used in
Factor Analysis tend to load on more than one factor suggesting their
importance to more than one factor - statistically a very discouraging sign for

factor analysis.

Regression Analysis Results for Theater Attendance

While the simple correlation analysis presented in the preceding section
provided some useful information about the relationships between the
independent and the dependent variables. the bivariate nature of simple
correlations does not allow one to assess the collective impact of a set of
independent variables on the criterion measure. As such. one of the more
sophisticated multivariate analyses. multiple regression. is used to better
understand how the independent variables collectively fit together to explain
the variance in theater attendance. In addition to allowing one to assess the
overall contribution of all the independent variables (the statistical fit. indicated
by R2). multiple regression also allow to assess the contribution of a particular
independent variable in the context of other independent variables (partialling

out the effects of other predictor variables).

108

Table 4.9 presents various summary statistics for the variables used in
regression and discriminant analyses. For those variables that are continuous
and polytomial. the means. standard deviations, and degree of skewness are
presented. For binary (dummy) variables, only the means are presented to be
used as proportions as standard deviation and skewness are not meaningful
with such variables.

The regression equations listed in Table 4.10 presents the results
obtained using the same set of variables but in different variable and sample
configurations to assess the models that best fit statistically yet are
theoretically meaningful. “Stepwise” method in SPSS is used after careful
consideration of the impact of other variable selection methods such as
“backward elimination" and “forward selection.” While no method is superior
than the other. it may be possible that one method may work better with one
sample than another (SPSS. 1990). Quite interestingly, all the three data
selection methods attempted (Stepwise. Backward and Forward) resulted in
practically the same statistical fit (R2) and more importantly. found the same
set of independent variables to be significant. As such. the more popular

“stepwise” method is chosen.

Equation 1. estimated using a partial sample of 301 cases as missing
values remain untreated. includes all 21 independent variables. Equation 2 is
estimated using all the variables in equation 1 but the missing values for the
variables INCOME (28 cases), MOVICOST (31 cases). HHHEADS (9 cases). and
PERCENTG (3 cases) were substituted by their respective averages. As seen
in the Table 4.10. the second equation. with the missing values treated.

provide a slightly better statistical fit as it explains about two percent more

TABLE 4.9

SAMPLE STATISTICS FOR VARIABLES IN REGRESSION AND DISCRIMINATORY ANALYSES

(N3366 UNLESS OTHERWISE STATED)

 

 

CORRELATION WITH
VARIABLE MEAN STD DEV SKEWNESS MOVIE# VIDRENT
MOVIE# 1.26 1.27 2.07 — .089
VIDRENT 4.12 3.95 1.80 .089 —
STARS 7.05 2.11 -105 .088 .007
DIRECTOR 3.77 2.53 .33 .204” .010
srommp 8.39 1.44 -137 .1089 .006
RATINGS 3.80 2.96 .45 .062 .033
REVIEWS 5.82 2.31 -103 .1299 .025
PREVIEWS 6.57 1.82 1.03 .091 .051
PRESTORY 4.79 2.27 -.07 .009 .006
WORDOFM 7.44 1.72 -134 .029 .001
AWARDS 4.41 2.42 .16 .175‘ .001
ADV'I‘SNG 5.74 2.00 -.76 .160‘ab .028
rvaomos 4.06 2.25 .04 .072 .044
SEQUEL 4.87 2.40 .014 .008 .035
PERCENTG 35.89 34.33 0.80 — -501“
MOVICOS‘I‘ 12.46 7.70 1.33 -.070 .267a
I-IHHEADS 2.89 1.57 1.15 .008 .319“
AGE 3.37 1.89 .35 -222“ -.135*
EDULEVEL 3.38 .90 .30 -.003 .134
INCOME 3.24 2.14 .57 -033 .036
BINARY VARIABLES

ADMSNS .75 na na .082 .121
MARITAL .41 na na -200“ .117
CHILDREN .29 na na -.111a .249“

 

*Significant at .01 1"Significant at .001
aBecame significant after mean substitutions bLost significance after mean substitutions

 

lIO

variance in the dependent measure (R2=.2069 vs R2=.l87l). However. the
second equation lost one of the independent variables (AWARDS) as the missing
value treatment change the original composition of the sample. Extreme care
was taken to see whether the missing values were random or take an
identifiable pattern. To find this out. the sample was first divided into two
groups--those observations with missing data on INCOME and MOVICOST (the
two variables with largest number of missing cases) and those with complete
data--and was examined the distribution of the dependent measure (MOVIE#)
using t-tests. The results found no statistically significant differences between

the groups thus confirming the randomness in missing values.

Equation 3 is similar to equation 2 but is estimated without the variable
INCOME. Although income was found to be a significant predictor for theater
attendance (as seen in equations 1 and 2). it appears to highly inter-correlate
with several other predictor variables suggesting potential multicollinearity
problems (see Table 4.7 - correlation matrix). The problem with collinear
variables is that they provide very similar information and may weaker each
other when used in conjunction. hence is difficult to separate out the effects of
individual variables. However. it is quite premature to exclude a variable as
collinear on the basis of simple correlation results as they might have different
effects in the presence of other variables. As such. two more sophisticated
diagnostic tests (Tolerance and Variance Inflation Factor (VIF) and Eigenvalues
and Condition Indexes) were performed using SPSS to see the true nature of
multicollinearity in the context of other predictor variables. As the results of
these tests suggest (see Table 4.11 in Appendix II for the test results) income.

in fact. seems to share an unacceptable amount of its variance with AGE and

TABLE 4.10

 

REGRESSION EQUATIONS FOR MOVIE ATTENDANCE

 

DEPENDENT VARIABLE: AVERAGE MONTHLY THEATER ATTENDANCE

 

NON STANDARDIZED BETA COEFFICIENTS

 

 

INDVARIABLE EQN 1 EQN 2 EQN 3 EQN 4 EQN 5
DIRBCIOR .0761' .1272’ .0903’ .1055‘ .1078‘
REVIEWS .0869' .0958’ .0963' .1231' .0194‘
ADVTSNG .0784' .0673‘ .0952‘ .0696‘ .0270‘
AWARDS .0543' .0843 .0610‘ .0537' .0319‘
STORY“? .0663 .0776 .0671 .0943' .0805
PREVIEWS .0611 .0516 .0740 .0735' .0396
RATINGS .0187 .0294 .0280 .0362. .0072
AGE 4354' -.1344' -1848" -- -.'0660
MARIIAL -6406‘ -6948' -.5610' - -2225'
INCOME .1557‘ .1749' — — —-
CONSTANT 1.07 1.13 1.63 1.39 -.188
R2 .1871 .2069 .1753 .1483 .2064
Amgz .1677 .1935 .1615 .1316 .1929
B (SIG) 9.63 (001) 15.47 (001) 12.71 (001) 8.90 (001) 12.11 (001)
STD ERROR 1.09 1.14 1.16 1.18 .372
N 301 366 366 366 366

 

EQUATION 2 MSING VALUES SUBsrrTUTED WITH MEAN VALUES

EQUATTON 3: RECRESED WITHOUT THE INDEPENDENT VARIABLE ” INCOME"
EQUATION 4: REGRESSED ONLY THE CREATIVE AND MARKETING VARIABLES

EQUATION 5: LOGARiTI-IMIC TRANSFORMATION (BASE 10) or THE DEPENDENT VARIABLE

t
SIGNIFICANT AT .05 OR GREATER
VARIABLES NOT SIGNIFICANT AT .05 LEVEL IN ANY OT THE ABOVE EQUATIONS: STAIS PREVIEWS, PRESTORY,

WORDOFM, TVPROMm SEQUEl, MOVICOST, ADhfihfi, CHILDREN, HHHEADS, EDULEVEL

 

l 12
MARITAL. Consequently. INCOME was excluded from subsequent estimates

and the results are presented in equation 3. The results. as it turned out.
seem to be a good compromise. Because. although equation 3 resulted a model
with three percent less variance explained than equation 2. AWARDS. which
lost its significance as missing values were substituted. re-entered the equation

as a significant predictor.

Equation 4 is estimated by regressing only the 12 variables that deal
with movie quality and marketing leaving the demographic and economic
predictive factors out. And this equation presents some very interesting
results. The new equation has a lower statistical fit as it only explains about
15 percent of the observed variability (R2=.l483). compared to the preceding
models. which is not unusual given the evidence of impact of other predictor
variables like AGE and MARITAL. Nevertheless. the new equation resulted in a
model with three new statistically significant theoretically sound variables
(STORYTYP. REVIEWS and RATINGS). Although the statistical fit is lower. this
equation provides some useful information relevant to the research question.
For example. as discussed earlier in this chapter. the importance of these
three variables significantly differs from group to group (ie., married to single,
“young” to “old.” and households with and without children) when group
differences were estimated using t-tests. Intrinsically. when the importance of
these variables was assessed collectively in regression analysis. the group
differences (impact of demographic variables) appear to prevent these variables
entering the equation as significant factors (as regression constantly assess the

interdependency of variables in its entry and deletion strategies).

l 13
The fifth and the final equation in this section. is similar to equation 3

but uses a logarithmic transformation (base 10) of the dependent measure.
The log transformation is a result of series of tests performed to check the
linearity of the continuous variables. A casewise residual test performed under
Subprogram Regression of SPSS revealed a slight departure of the dependent
variable from normality (this may be due to the way in which values were
assigned to the dependent measure. described early in the preceding chapter.
which created a large number of cases for two values). Under these
circumstances. SPSS suggests the transformation of either the dependent or
independent variables. or both. to approximate normality (SPSS. 1990). All
three transformations were attempted and only the transformation of the
dependent measure resulted an improvement in the statistical fit. As the
distribution of the variable has a positive skew. the log transformation appears

to be the most appropriate.

Quite interestingly. the equation with the transformed criterion measure
achieved a statistical fit equal to that of the equation 2 even without the
INCOME factor. In terms of the number of variables in the equation. both.
equations 3 and 5. have the same variables while the latter has a better
statistical fit of almost 21 percent of the variability explained (R2=.2064) when
the dependent measure is “straightened out” with log transformation. Thus

making it a better model.

Some other interesting things happened when we move from the simpler
to more complex statistical analysis. For example. while CHILDREN is
negatively significantly related with movie attendance in correlation analysis. its

significance disappeared in combination with other variables and did not enter

114

into any of the equations. It may be possible that the children factor can
works both ways. Although having young children put an additional burden on
going to the theater (as it is necessary to find a sitter and other logistics). the
children that are relatively older may be a factor that attracts parents to the
theaters. thus offsetting the negative impact. This is also evident from the
responses given to one of the open-ended items where 7.7 percent cited “kids

wanted to see” as the reason for going to the last movie.

However. RATINGS. PREVIEWS. and ADVTSNG. which were not
significant in correlation analysis. but appear to share their variances with
demographic factors. did enter into the model in the absence of demographics
(equation 4). What is quite noticeable is the entry of more or less the same set
of variables in all equations under difi'erent sample and variable configurations.

This indicates the robustness of the estimation process.

In looking at all the five equations collectively. there are four creative
variables. DIRECTOR. REVIEWS. ADVTSNG. and AWARDS. and two
demographic variables. AGE and MARITAL. that always relate significantly to
the theater attendance. This is true even when the sample and the variable
configurations were altered. indicating their strong association with the

dependent measure.

In all five models. the entire set of independent variables is linearly
related to the dependent variable as all the F values are all significant at .001.

The F is a test of the null hypothesis that 31:32:33=34....Bn=0. The statistical
fit (R2) ranges from .1483 to .2069 meaning between 15 percent to 21 percent

of the observed variability in theater attendance can be explained by

DIRECTOR. REVIEWS. ADVTSNG. AWARDS. STORYTYP. PREVIEWS.

1 15
RATINGS. AGE and MARITAL. While there is no such thing as good. best or

optimal R2. it can be viewed as a measure of the explanatory power of a
regression equation. Provided theoretically sound variables are included. R2 is
one statistic that can be used to evaluate the “goodness” of an estimated
regression equation (Johnson. Johnson and Buse. 1987). However. as R2 is
greatly affected by the data sources used to estimate the regression equations.
given the use of individual household data. 15 to 21 percent of explained
variance provide encouraging and acceptable results. Moreover. this range of
explained variance is consistent with other communication research at

household level.

Discriminating the “Frequent” and “Infrequent” Moviegoer

In an effort to further explore data beyond multiple regression analysis.
another linear statistical technique. discriminant analysis. is used where the
same sets of independent measures serve as “discriminatory function
variables” to classify cases into one of the groups. As such. the same criterion
variable. average monthly movie attendance (MOVIE#), is converted into
discrete form to divide sample into two groups called “frequent” and
“infrequent.” The “frequent” category included those who go to the movies at
least once a month and “infrequent” category included those who attend movies
less than once a month. There were 127 (41%) “infrequent” moviegoers and
184 (59%) “frequent” moviegoers in the sample (N=31 1). In fact. this form of
the dependent measure is more close to its original form than the one used in
regression analysis. As the assumptions for discriminatory analysis remain
similar to those for the multiple regression. INCOME was not considered as a

sound predictive factor in the analysis.

1 16
Table 4.12 presents the results of discriminant analysis. Standardized

canonical discriminatory function coefficients (similar to the beta weights in
multiple regression) are presented along with the means and standard
deviations for each variable and other summary statistics. Prior probabilities
were set at .41 for “infrequent” group (as 41% of the sample belong to this
group) and .59 for the “frequent” group reflecting the actual proportions. The
overall model is significant at oc =.001 (X2=72; df=9). meaning the mean
differences were not due to chance alone. Wilks Lambda is .7891 and
canonical correlation is .4592. As the variation explained in the dependent
measure by the discriminatory function variables equals one minus lambda (l-
.7891=~21%). the smaller the lambda the better the statistical fit of the model.
In two group situations (eg. Frequent vs Infrequent). lambda is equal to the
squared canonical correlation coefficient (.45922=~21%) which is similar to R2
in regression analysis. The statistical fit of the discriminatory model. therefore.
is similar to what is achieved with the best regression model.

Table 4.13 present the classification results of discriminatory analysis.
Correctly classified cases appear on the diagonal of the table (bold faced). For
example. of the 127 cases in the “infrequent” category. 83 (65.4%) were
predicted correctly to be the members of that group while 44 (34.6%) were
“misclassified” to the infrequent group. Similarly. 133 out of 311 (72.3%) of
the frequent group were correctly identified while 51 (27.7%) were
misclassified. Overall. the discriminatory model correctly classified 69.45
percent of the cases (216/311). The combined distribution of the scores for
the two groups is shown in Figure 4.1. One symbol represent five cases and

the overlap between the two groups can be seen by the overlapping symbols.

TABLE4.12

 

DISCRIMINANT ANALYSIS OF MOVIE ATTENDANCE

 

DEPENDENT VARIABLE: AVERAGE MOVIE ATTENDANCE CLASSIFIED AS
"FREQUENT" AND "INFREQUENT"

 

 

MEAN (STD DEv)

Dsc FUNCTION STD COEFF INFREQUENT FREQUENT
DIRECTOR 1650' 3.56 (2.45) 4.11 (2.58)
Swamp .1435‘ 8.45 (1.31) 8.39 (1.52)
REVIEWS 3167' 6.07 (2.07) 5.58 (2.46)
AWARDS .3152” 4.00 (2.18) 4.73 (2.55)
ADVISNC 1964* 5.40 (2.29) 6.00 (1.78)
AGE -.8717* 4.02 (1.83) 2.75 (1.68)
MARITAL -2743" .48" .31“
ADMSN$ .2883' .66“ .84”
mm -.1705' .37“ .49“
CONSTANT 1.298

 

'SIGNIEICANT AT .05 OR HIGHER
X 2=7200 DF=9
WILRS LAMBDA .7891 — CANONICAL CORRELATION .4592

VARIABLE NOT SIGNIFICANT AT .05: ST ARS RATINGS, PREVIEWS, PRI‘B’TORY, WORDOFM,
TVPROMOS, SEQUEL, MOVICOST, HHHEADS, EDULEVEL

H
STANDARD DEVIATIOBs NOT MEANINGFUL

 

Consequently. if the discriminatory (independent) variables used in the model
were able to predict the group classification in total accuracy. there would be
no overlapping symbols in the graph. In other words. the overlapping shows
the “unexplained” variance in the criterion variable. The correct classification
percentage of 69.45% and the statistical fit of 21% seems reasonable for non-

aggregate household data.

118

TABLE 4.13

 

CLASSIFICATION RESULTS: DISCRIMIN ANT ANALYSIS

 

PREDICTED GROUP MEMBERSHIP

 

ACTUAL GROUP N INFREQUENT FREQUEN'T
INFREQUEN'I‘ 127 83 (65.4%) 44 (34.6%)
FREQUENT 184 51 (27.7%.) 133 (72.3%)

 

PERCENT OF "GROUP” CASE CORRECTLY CLASSIFIED 69.45% (216 / 311)

 

 

 

 

 

 

 

 

FIGURE 4..1
— —
ALL-GROUPS STACKED HIST OGRAM
CANONICAL DISCRIMINANT FUNCTION
40 + +
2
2
30 + 2 +
2
2 2
2 2
20 + 22222222 +
22222222
2222222222
22 2212222222
10 + 12211122122222 +
2 1 12211122122222
2 1 11111111111111222 2
11 111111111111111111111222 2
x----+-- + ———+—— + —— 4 — 4 —+— x
Out -3.0 -2.0 -1.0 .0 1.0 2.0 3.0 Out
Class 11111111111111111111122222222222222222222
Centroids 1 2
EACH SYMBOL = 5 CASES 1=INFREQUENT; 2=FREQUENT

 

 

 

 

119

Results of the discriminant analysis appear to be theoretically stronger
and statistically similar to regression analysis. Five creative and marketing
variables (AWARDS. REVIEWS. ADS. STORYTYP and DIRECTOR). together
with age (AGE). admission prices (ADMSNS). marital status (MARITAL). and
children in the household (CHILDREN) are statistically significant
discriminatory variables for correctly predicting the “frequent” and “infrequent”
moviegoers. It is certainly an improvement over the regression analysis to see
five creative and marketing variables entering the equation together with
demographic factors. which did not happen in any of the regression equations.
In addition. the significance of two other predictor variables. CHILDREN and
ADMSNS. also shows further improvements as they provide more insight
information. How and why the discriminant analysis resulted a better model

with same data will be discussed in the “discussion” section.

4. 1 0 Regression Analysis Results for Video Rental

The regression equations listed in Table 4.14 presents the results for
video rental patterns. The same set of independent variables. plus a new
predictor variable. labeled PERCENTG. are used to estimate these equations
where the dependent variable is the average monthly video rentals (VIDRENT).
The new predictor variable, PERCENTG. represent the proportion of movies
seen in the theaters over the past six months. Income is also a significant
factor in video rentals but was excluded from subsequent analysis due to its
known problems. As noted elsewhere in this chapter. the theoretical
justifications for the inclusion of these variables are coming from their usage to
explain theater attendance -- consumption of the same product (i.e.. movies)

but through a different medium.

120
Normality test performed on the dependent variable (VIDRENT) revealed

to have a fairly normal distribution therefore no need for transformation. Entry
techniques and assumption checks used in these equations are similar to the
ones used in theater attendance analysis. Similar to the previous analysis.
three equations are estimated. one without treating the missing values and the

other with missing values substituted with mean values.

TABLE 4.14

 

REGRESION EQUATIONS FOR VIDEO RENTAL

 

DEPENDENT VARIABLE: AVERAGE MONTHLY VIDEO RENTAL

 

NON-STANDARDIZED BETA COEFFICIENTS

 

 

INDVARIABLF. EQN 1 EQN 2 EQN 3
AG! -5331‘ -3224' -4910"
Pam-mo .0485" -0465' -.0504*
WW 1962' .2655‘ .3052‘
MOVICOST .0948. 0599' .0726
INCOME .2950' — ——
CoNSTANT 6.25 5.36 6.90
R2 .3210 .3003 .2746
ADIRZ .3094 .2916 .2665
F (SIG) 27.70 (001) 27.87 (001) 33.87 (001)
STD ERROR 3.14 3.12 3.39
N 301 301 366

 

EQUATION 2- WTTHOUT INCOME, MESING VALUES UNTREATED

EJUATION 3: WITHOUT INCOME, MEING VALUE TREATED

t
SIGNIFICANT AT .05 OR GREATER

VARIABLE NOT ENTERED TO AT LEAST ONE OF THE EQUATIOBE ALL 12 CREATIVE
AND MARKETING VARIABLE, ADABN5, HHI-IEAIB, EDULEVEL

 

12 1
Equation 1 in table 4.14 shows the five variables that are significant

when all the variables. including income. were entered using the “stepwise”
method. The equation is estimated using a partial sample (n=301) as missing
values remain untreated. Equation 2 is estimated without income and equation
3 with same variables as equation 2 but with missing values substituted with
mean values. As seen in table 4.14. once the missing cases were included.
MOVICOST lost its significance and did not enter the equation. It also lowered
the explained variance by about 2.5 percent. All the other variables are still
significant but no new variable entered to the model. In a situation like this it
is advisable to keep the previous model as the new model is weaker in
theoretical power and the statistical fit than the previous one. The actual

remedy here is to have a larger sample. which is not practical at this point.

What is obvious at the outset is the insignificance of all the creative and
marketing variables in all three equations. Although the extremely low
correlation coefficients between VIDRENT and all the creative and marketing
variables had forewarned this outcome. it is surprising to see that none is
entered as significant. The only variable that came close to the entry limit
(significant at .06) is SEQUEL. a variable that was not significant with any of
the theater attendance equations but makes sense as the audiences may want

to see the prequel on video.

Age still remains as a significant factor in explaining video rentals.
However. it is somewhat surprising to see a negative coefficient for AGE
because it seems reasonable to expect people relying more on the convenient
home video window as they grow older. Unlike in the movie attendance

equations. marital status, however. is not a significant predictor for video

122

rentals. This seems quite logical as watching a video takes place at home. thus
eliminating the need to go out. which is more appealing to the married. As
expected. PERCENTG has a negative coefficient. suggesting a relatively low
tendency toward video rental among those who attend theaters more.
CHILDREN entered to the video rental equation with a positive coefficient
suggesting the households with children tend to rent more than non-children
households. The significance of MOVICOST also suggests that for those who
have to spend more for a movie outing. home video provides a less expensive
alternative. Again. the importance of both of these variables is not surprising
as the early results found that “price” and “convenience“ to be overwhelmingly

attractive reasons for renting rather than going to the theater.

The statistical fits (R2) of theses’ equations range from .2746 to .3210
which is about 10 percent more than the equations estimated to explain the
theater attendance. The most important variable in explaining the Video rental
is PERCENTG which alone explains about 17 percent of the variance in video

rental.

In the following chapter these results will be discussed and some

conclusions will be drawn upon with suggestions for future research

Chapter 5
Discussion and Conclusions

In this chapter. the results reported in the previous chapter will be
discussed and conclusions will be drawn upon. Suggestions for future

research will also be presented in the light of present research experience.

Theater Attendance

The average adult goes to the movies about once a month. If the
industry classifications were to be used. about one-half of the respondents are
“frequent” moviegoers. To a large extent. the sample statisch closely match
the attendance habits found by the industry research and some prior studies
(Austin. 1989: MPAA. 1992). For example. age. education. and marital status
significantly influence the theater attendance habits as young. single and the
college educated tend to go to the movies more often than their respective
counterparts. Consistent with what the industry has noted lately. there seems
to be an upward swing in attendance among older demographic segments. For
example. the average attendance rate of those who are 50 to 59 years of age
ranked only behind the young adults (16 to 24 age category). However. as the
differences did not attain statistical significance. it seems premature to
conclude that whether this is a trend or just an aberration. Likewise, the
tendency among males to frequent theaters more often than females shows no

statistically significant regularity as well.

123

124
Besides demographics. only the subscription to movie channels (pay

cable) seems to have a significant influence on movie attendance. Quite
interestingly. however. watching movies on premium channels apparently is not
a substitute to theater attendance. Instead. those who subscribe to at least
one movie channel tend to go to the movies significantly more than those who
do not. suggesting that it may be their interest in movies that might have
motivated them to subscribe to movie channels in the first place. This
relationship may also help explain why the annual attendance figures remained
largely unaffected over the past ten years in spite of the growth of these
alternative exhibition windows. The following section will focus on the specific

factors that influence the attendance decisions.

Determinants of Theater Attendance

In looking at the results of collective impact of predictor variables on
theater attendance several findings are of note. In all five regression equations
five variables. DIRECTOR. ADVTISNG. REVIEWS. AGE. and. MARITAL were
always significantly related to attendance even under different sample
conditions and variable configurations. AWARDS was significant only under
different sample size. STORYTYP, PREVIEWS. and RATINGS attained
significance only when creative and marketing variables were entered without
demographic variables. Among all the demographic variables included. AGE

and MARITAL were the only to attain significance.

Age emerged as the most important predictor of attendance and the first
variable to enter the equation explaining about 6 percent of the variability. The
negative coefficient of AGE suggests an inverse relationship between age and

theater attendance and confirms the known association that exist between the

125
young and movie attendance. The strong relationship between age and

attendance is also evident from the group t-test results. As bivariate analysis
indicates. the young moviegoers not only dominate the attendance. but are also
quite indifferent to the quality of the movies in their selection process. For
example. factors such as the story type. critical reviews and ratings seem to
matter less for them than to the older audiences. Instead. their attendance

decisions are largely influenced by advertising and promotional factors.

Marital status. a binary variable that is significant in all equations. was
able to add 4 percent to the explained variance. The negative coefficients of
marital status that ranged from -.5610 to -.6948. tells us that being married
decrease the number of movies seen by .56 to .69 than their single
counterparts. Similar to age. marital status is also associated with other
factors that influence attendance decisions. As group t-test results revealed.
the married audiences place a significantly higher value on the factors such as
the story type. word-of-mouth advertising. awards and ratings. in their

decisions to attend movies.

The importance of advertising and promotional factors on attendance
seems overwhelming. Of the six advertising and publicity related variables
included (ADVTISNG. REVIEWS. PREVIEWS. AWARDS. WORDOFM. and
TVPROMOS). four variables were significantly related to attendance decision.
The most prominent are the paid advertising in the media and critical reviews.
In all the equations estimated. advertising is the most important variable
besides age. In terms of the explanatory power. advertising is equal to marital
status that added 4 percent to the explained variance. The significance of

critical reviews is consistent and added about 2 percent to the statistical fit.

126
The AWARDS (nomination or winning of major awards) also a significant

contributor to attendance decisions as it entered all but one equation. Theater
previews. a new variable which was adopted from Austin’s model. is as
important as reviews but its entry was conditional. Previews entered as a
significant predictor of attendance in the absence of demographic factors
suggesting its appeal to the specific demographic sectors. The insignificance of
word-of-mouth is quite surprising given its reputation as the most effective
form of “advertising” for movie attendance and the rank it earned as the second
most important factor affecting attendance decisions (see Table 4.2). Another
new promotional variable attempted for the first time. TVPROMOS. (appearance
of movie stars on television programs like Tonight Show or the Late Show) did
not gain significance in any of the equations. Perhaps its popularity is not yet
widespread or established or may not have been measured comprehensive

enough.

Despite their wide recognition as important factors affecting attendance.
this is the first time practically all advertising and publicity related factors were
included in a multivariate predictive model to explain theater attendance.
Consequently. the significance of four advertising and publicity related
variables uncovers some new information and confirms some of the early
findings of communication and economic research. The strong relationship
between attendance and advertising. together with the significance of theater
previews. provide empirical evidence to show filmgoers’ reliance on outside
sources for information to help them with decision making. The regression
results. together with the results of bivariate analysis. show that advertising

and promotional factors have an overwhelming impact on attendance decisions.

127

It is no wonder then the industry is spending as much to advertise and
promote a movie as to make one. The advertising budgets. for example. have
more than doubled to an average of $12 million over the past ten years--
apparently to fulfill the moviegoers' thirst for information. As Litman and Kohl
(1989) noted. advertising serves a dual purpose. First it lures the avid
moviegoer to the theater and then it helps build the word-of-mouth campaign.
Current data suggest that the advertising and promotional monies are well

spent and help meet the aforementioned goals.

Overall. the creative variables fared very weakly in regression equations.
For example. from the five creative variables included (STARS. DIRECTOR.
STORYTYP. SEQUEL. and PRESTORY). only the director entered in all five
equations. Story type (genre) attained significance only in the absence of
demographics. Each of these two variables added 2 to 3.5 percent to the
explained variability in attendance. The significance of DIRECTOR seems more
consistent as it entered in all five equations suggesting the familiarity of the
director does matter in attendance decisions. Quite interestingly. the
importance of the director did not gain much support in individual ranking of
important factors on attendance. Yet. director attained significance in all
equations. On the contrary. STARS was ranked as the third most important
factor in attendance decisions and was the most cited reason that attracted
filmgoers to their last movie. Still. STARS didn't attain significance in any of
the equations. Perhaps. what's a star is too subjective or abstract to measured

or defined.

128

Distinguishing “Frequent” and “Infrequent” Moviegoers

The discriminant analysis was performed with the desire to statistically
distinguish between the “frequent” and “infrequent” moviegoers using the same
set of independent variables as discriminating factors. It was believed that the
operational procedure used to measure the dependent variable in regression
equations (average attendance level) would probably be better if treated as a
discrete variable than a continuous one. The erratic nature of movie
attendance by a large number of respondents-- 167 of the 366 surveyed go to a
theater ‘once about every other month’ or “once about every six months'--further
supported this decision. The dependent measure was then converted to a

discrete variable to form “frequent” and “infrequent” groups.

The results of discriminant analysis are statistically similar to the
regession results as it also explained about 21 percent of the variability in the
dependent measure. The model also correctly classified 69.45 percent of the
cases using nine independent variables of the model as discriminatory
functions. The variables in the discriminatory model. it generally confirm the
regression results rather than breaking new grounds. Similar to the regression
equations. age entered as the most important discriminatory variable
contributing most to the overall discriminant function. Similar to the regression
models. demographic and advertising/publicity variables emerged as stronger
contributors to the overall discriminant function than the creative factors.
There are. however, some minor improvements in the discriminatory model
compared to the regression analysis. First. the model was able to correctly
predict “frequent” and “infrequent” moviegoers using five creative and

promotional variables together with demographics. As such. the demographic

129

variables seem to cooperate with creative and publicity factors to distinguish
group differences. Second. two new variables. having children in the
household and admission prices. emerged as significant predictors of
“frequent” and “infrequent” moviegoers. The negative coefficient of CHILDREN
suggest that those in the households with children are less likely to be in the
“frequent” moviegoer category. The significance of the ADMSNS suggest a
higher probability of falling into the “frequent” moviegoer class if admission
prices were lower. However. it seems that the most significant contribution of
discriminant analysis is the reconfirmation of the regression results. The
minor improvements are the benefits one gains by compromising the

dependent measure to a “cruder” discrete form.

Overall. the findings emphasize the overwhelming importance of
advertising and promotional factors on attendance decisions. Their influence is
strikingly greater on the younger moviegoers but all demographic subgroups
seem to be receptive to advertising and promotional campaigns. These
empirical findings support the importance of Austin's “Publicity 81 Advertising
Focusing Ring” in his “Zoom Model.” hrrthermore. the significance of director.
critical reviews. awards. and story type. are consistent with the results of
economic/business studies of Litman. (1983) and. Litman and Kohl (1989).
The degree of consistency found between the current research and
economic/business research is encouraging given that they were arrived at
using entirely different data sources and different research methods. The
importance of demographic factors needs to be evaluated beyond the
significant entry of age and marital status into regression equations. The

finding that some creative and publicity variables (story type. theater previews

130
and ratings) attained their significance in the absence of demographic variables

suggests the interdependence of these variables with demographics and their
relationship with attendance. The results showed that demographic
differences not only have direct effects on attendance. they also affect the
significance of other factors and in turn. influence attendance decisions. In
the empirical world. these complexities are translated into the appeal of certain
aspects of the movies to certain groups and the importance of social
institutions in attendance decisions. More sophisticated research designs are

needed to sort out these complex relationships.

Video Rental

While the results discussed in the preceding sections tend to help
explain movie attendance and reconfirm some of the early results. the most
revealing findings are the ones that shed light into the viewing that takes place

outside the traditional theater--which will be the focus of this section.

Among the three leading delivery channels that compete with each other
to bring movies to living rooms (pay cable. pay-per-view. and home
videocassette). home video has emerged as the indisputable winner. Despite
the heavy concentration of young. educated and high income households in the
sample. popularity of both. pay cable and pay-per-view. seems minimal. Less
than one-fourth of all the households subscribe to at least one pay cable
channel and. even among the cable subscribers. the subscription rate for
premium channels only goes up to 29 percent. Pay-per-view even fared worse.
Although it is largely accessible. with availability to over 80 percent of the
households. only 14 percent of them have ever ordered and paid for a movie

on pay-per-view. The lukewarm acceptance of pay cable and pay-per-view

131

seems to justify the recent industry decision to give home video the exclusive 4
to 5 week window before making the movies available to pay cable and pay-
per-view. It appears that the industry is simply. and correctly. responding to
the signals it receives from the market to maximize profits by “determining the
optimal sequencing and time clearance for each successive exhibition window”

(Litman and Kohl. 1989).

With an 86 percent penetration rate22 and average rental of over four
cassettes per month. home video is undoubtedly the most widely used
exhibition window for the movies. Eighty-five percent of the households with a
VCR rent two or more tapes a month and virtually all VCR households rent at
least one movie per month. According to industry classifications. these
statistics means that VCR has turned more than 80 percent of the nation‘s
households into “frequent” movie viewers. A technology that started as a time-
shifting device for the avid TV viewer had eventually rose into the pinnacle of
movie offering. The mere size of the home video audience speaks for this. For
example. the average audience of 2.9 persons per each video rental”. taken
together with the industry estimates of 4.1 billion videos rented last year. put
the audience size for home video over 10 billion mark. This is more than
double the size of the audience the movie industry had enjoyed during its
heydays in the late 19408 when weekly attendance levels reached 90 million
and several times greater than the annual 1.1 billion attendance marks for

theaters for last three decades.

 

22Close to the 88% penetration rate the industry had predicted by the year 1995.

23This is the weighted average after dLscounting for possible fraternities and sororities in
the sample by trimming the top 10% of the cases. The sample average is 3.3 per/HH.

132

As evident from these results. home video’s huge success can be largely
attributed to two factors: price and convenience. Audiences perceive home
video as the least expensive and most convenient window to see movies. But
price and convenience are only half the story and tend to overlook the role
played by the industry to bring about home Video to its current popularity. It
is the classical “invisible hand” that put the private interest of the industry with
the interests of individuals to create the current situation. For example. as the
financial significance of home video increased. the industry responded by
gradually shortening the clearance time between theater and video-~sometimes
to a point where the video is released when the movie is still in its first
theatrical run--and shifting the clearance sequence from pay-peT-View to home
video. As a result. home video has become the first exhibition window outside
the theater to offer movies in such a short time to more than 80 million of the
nation's households. Neither pay cable nor pay-per-view has given such an
appealing opportunity to such a large populace before. Moreover. unlike pay
cable or pay-per-view. home video does not select titles: each and every movie
is guaranteed to come on video sooner. not later. The ones that did well in the
theaters come sooner to reap the benefits of successful theater performances
and the word-of-mouth publicity. The ones that did not do well also come
sooner to recoup whatever it can from the video market before it completely
fades away from the people’s mind. As a result. about one-half of the audience
now “always” or “almost always” wait for the video to see the movies.
Furthermore. more than one-half the audience would rent the video rather than
going to the theater if the option were Offered to them by simultaneously

releasing movies to theater and on video. And a majority is willing to pay more

133

for such an early rental opportunity. The specific factors that influence video

rental will be the focus of the following discussion.

Detemiinants of Video Rental

Quite surprisingly none of the twelve creative or promotional variables
included in regression analysis are significantly related to the video rental
patterns. However. the insignificance of marketing related variables can be
attributed to the virtual absence of advertising efforts to promote video releases
by the movie industry. In fact. information on new video releases is only
publicized through in-store advertising and on videos themselves (similar to
theater “trailers”) which only exposes them to the already committed renter.
Only Disney titles and “mega-hits” that are meant for sell-through market use
some TV advertising and product tie-in promotions. Still. they are quite limited
and nominal compared to the big promotions for the initial theatrical release.
As such. it probably is the absence of easily identifiable advertising and

promotional strateges that made these variables insignificant.

The insignificance of creative variables such as. the STORYTYP. STARS
or DIRECTOR depicts the indiscriminant renter who appears to relax his or her
standards when renting a movie compared with going to the theater to see one.
Naturally. the relatively lower financial risks involved with renting (the average
rental is around $2 per video compared to the average $5 admission per
person plus concession prices) and the low involvement associated with renting
(logistics involved such as. finding a sitter. leaving home) seem to explain why
the renters apparently place a lower value to creative factors when it comes to
renting. In fact. these findings suggest that the moviegoers may have

developed two different standards for the two exhibition windows. The ‘try

134

anything’ attitude for the videos and ‘play cautious‘ attitude for going to the
theaters. conceivably based on the cost and involvement. A bad rental choice.
therefore. is financially more tolerable and the “Opportunity cost” can be
minimized by turning the VCR off--much easier task than leaving the theater
especially when the movie is often the center of the planned activities for the

evening.

The four variables that are significant seem to clearly explain the
dynamics of the home Video market. Percentage of movies seen in the theaters
(PERCENTG) emerged as an important predictor of video rental which alone
explained about 17 percent of the variance in video rental. As expected. the
negative coefficient of the percentage suggests an inverse relationship between
theater attendance and video rental. This relationship appears to indicate that
home Video. unlike pay cable or pay-per-view. is gaining acceptance from the
moviegoers as a substitute to theater attendance. an indication that suggests
possible audience shift from the theaters. However. the use of this variable as
a predictor of video rental should be done somewhat cautiously as this seems
to be an endogenous variable. The strength of PERCENTG may be overblown
to an extent since all the respondents that do not attend movies but do rent

them were given 100 percent.

The second variable to enter the equation was MOVICOST--the average
cost for a movie outing. It added about 5 percent to the explained variance
and is positively related with video rental suggesting that rental is more popular
among those who have to spend more at the theater. Naturally families with
children fall into this category. As correlation results indicate. households with

children also go to the movies less but rent more videos. Therefore. as

135
expected. CHILDREN is significantly positively related to video rental. The

significance of CHILDREN suggests that home Video serves as a better

(cheaper and convenient) way to see the movies for families with children.

The significantly negative relationship between age and video rental is
somewhat surprising given its similar relationship with theater attendance. It
seems logical to expect a positive relationship that suggests a higher rental
tendency among the older people as renting is more convenient than theater
attendance. However. if home video is emerging as a substitute for the theater
window for all the moviegoers. the negative relationship appears to be the
correct direction. However. further research with larger. better representative
samples will be needed to find the proper direction of the relationship between

these two factors.

What are the implications of this enormous success of home video on
theater attendance? The results of this study show that home video has
greatly enhanced the accessibility to the movies and consequently increased
the net audience to an unprecedented level. While it is almost impossible to
dissect the video audience into its root components. the data. however. seems
to suggest some identifiable trends. First. it is clear that the moviegoers did
not totally desert the theaters to flock the video stores. It is more likely that
moviegoers became more selective in their theater attendance in the face of
video alternative. As such. it is likely that the growth of video audience will
eventually come out at a cost to the theater. However. it will be a while until
the true impact of the audience erosion might show up in attendance figures
and a link can be established. The process would also be a rather slow one as

neither the industry nor the moviegoers apparently want to totally abandon the

136

theater. Yet. as the industry continues to shorten the clearance time between
theater and video as a measure of profit maximization and viewers. on the
other hand. continue to acquire better viewing facilities (stereo VCR. big screen
TV). as measures to optimize the benefits. home Video is likely to affect the

theater attendance levels that remained quite stable since the 1960s.

Interestingly. unlike the early days of TV. when the industry fought it
tooth and nail to hold the audiences defecting to TV. the current shift seems to
be carefully orchestrated by the industry through timely clearance and proper
sequencing of home video releases for the obvious reason that it is not losing
its audience to another medium. In fact. a financially strong home video
window. which appears to have no strong seasonal bias like the movies. will
have a tendency to solve some complex distribution problems as studios may
be able to avoid fiercely competitive release dates of summer and Christmas.
Finally. after everything is said and done. the distributors will be the ultimate
winners as the new exhibition windows constantly enlarge the net audience for

the movies. A significant change for a industry that enters its second century.

Limitations and Suggestions for Future Research

One of the great strengths of this study appears to be the timeliness of
the research questions it attempted to address at a time when the movie
industry seems to be at crossroads. The proliferation of new technologies.
especially home video. is bound to change the way motion pictures are
distributed and exhibited. and above all. consumed. This timeliness also
presented some obstacles that need to be addressed in future research to fully

understand the overall movie attendance habits.

l 3 7
The most important aspect at this point seems to be the development of

a comprehensive research plan to fully explain the home video industry. As it
was evident from the current research. our knowledge about the home video
industry seems limited to the annual rental figures provided by the industry.
Even from the supply side. no information is available at least to know whether
there are seasonal tendencies or demographic/psychographic differences exist
with video rental patterns as it is the case with theater attendance. This
dearth of knowledge directly affects the hypothesis development and the
development of measuring instruments. As such. the primary task of the
future research in this area is to carry out a comprehensive exploratory study
of the role of new technologies (specifically home videocassette. pay cable and
pay-per-view). to understand the uses of these exhibition windows. In
addition. there should be a parallel survey of the supply side of the industry as
the industry appears to take proper cues from the market behavior. The latter
would be more demanding and painstaking as it involved proprietary data the
industry prefers not to share. Yet. careful reviews of trade journals appear to
fill much of the gaps as more information seems to emerge with the maturing of
these technologies. Another possibility appears to be a collaborative research

effort with the industry.

In addition. there are some methodological concerns that need to be
addressed pertinent to this line of research. First. the future research should
begin with a larger. nationally representative sample (in effect. several
nationally representative markets). This seems quite important at the
exploratory level since future hypotheses will be derived based on this

benchmark research. Moreover. the penetration and the uses of other

138
exhibition windows. such as PPV and pay cable. tend to show their true uses

in a larger national sample. Although the current study used a random sample
of the general public. the biases associated with an atypical market like

Lansing were painfully apparent in the data analysis stages.

It also seems impossible to study both. theater attendance and Video
rental using one theoretical and one research approach. Furthermore. as home
video maturing to a separate market. a whole new theoretical approach seems
quite necessary to study this secondary market. Conventional predictor
variables used in past research to explain theater attendance seems no longer
sufficient for understanding video rental patterns. Consequently. it seems
more appropriate at this point to suggest separate demand estimates for each
sector of the industry (i.e. theater attendance and video rental) with different
predictor variables based on different theoretical models. The boundaries of
this line of research should also be expanded to include the audiences that
watch movies on over the air TV. While survey research provides very useful
information to measure attitudes and behavioral patterns. it seems necessary
to supplement this method with some qualitative methods such as depth
interviews and case studies that might also include the study of groups of
renters in selected markets. ideally with the cooperation of some video rental

stores.

Finally. it seems quite necessary that future research should benefit
from both. communication and economic research approaches in
conceptualization and operational stages as no singly theory seems to explain
this complex consumption pattern totally. First. as evident from the results.

convergence of communication and economic approaches would allow the

139

inclusion of structural changes into the research design together with changing
movie consumption patterns. This seems quite necessary as home video now
seems to emerge as a substitute to theater attendance. These changes in the
demand side have drastically altered the decision making process of the supply
side of the industry which is evident from the way movies are sequenced and
given time clearance for each successive exhibition window thus making the
inclusion of the economic approach almost mandatory in future analysis of
movie attendance. The necessity of a hybrid approach is also evident from the
significant influence found in demographic variables in the predictive models
for movie attendance and video rentals. In the present study. for example.
only the demographic factors emerged as significant predictors of video rental
patterns demonstrating their influence as well as the need for different
predictor variables in future research than what was used in the current
research. As noted before the use of conventional predictor variables
employed to explain the theater attendance seems no longer adequate to
explain the growing home video usage. Furthermore. some variables. such as
the effects of word-of-mouth advertising and psychographic factors can be
more appropriately be measured from the demand side using the “self-reports"
of the audiences. Above all. it is also evident that as the movie industry grows
into a multi-facet industry with several alternative windows and moviegoers
seems to have developed different standards for different exhibition windows.
As a result. no single approach or research method would be sufficient to
single-handedly understand the nature of the industry and the changing
audience needs. It is in this context that a convergence of two research

approaches seems quite beneficial in future research.

Appendix 1

n In

E

140

Hello. my name is , and I am calling from the Department of Advertising at Michigan
State University. We are conducting a survey to help understand the movie attendance and video
rental patterns of the general public. This survey will take about 6 to 8 minutes to complete and no
one will try to sell you anything. The information you provide us will remain confidential and
anonymous. Your participation in this survey is entirely voluntary and you may choose not to answer
certain questions or may discontinue the interview anytime. Are you 18 years or older?

[IF "YES” (DNTINUE IF NOT ASK FOR AN ADULT AND REPEAT THE INTRO]

To begin with... Have you seen a movie in a theater or rented one during the past six months?

Dves¢ GOTOQ#1 DNOCO THANKTHERESPONDWANDTERMNATE

 

1. In an average month about how many times do you go out to see a movie in theaters?
[consider only the theater attendance]

0 1 2 3 4 S 6 7 8 9 10+

[If “0” ask whether the respondent go to the movies at least ‘once about every two to six
months’ or ‘about once a year?’]

22 [3 Once about every two to six months
33 D About once a year
99 [3 DK/RF 88 C] Never =9 SKIPTOQ#5

2. Can you recall the last movie you saw in a theater? 2 C] NO :9 SKIP TOQ#4

1C3 YES ‘1' [WRITETHENAMEBELOW] 9C] DK/RF

3. What attracted you to that movie?

 

 

4. With whom do you usually go to the movies?

1 [3 With friends

2 C With acquaintance
3 C] With family

4 (3 With a date

5 D Alone

6 C] Other

9O DK/RF

':JI

{\

{L

11

11

I]

II

It

141

5. Now I am going to read you some of the factors people say are important to them when they
decide which movie to see. On a scale of 0 to 10, where 0 means “not important at all” and 10
means “very important.” please tell me how important each of these factors is to you when you
decide which movie to see. How important is...

Q that the movie has a star or stars you like?

Nor WANT AT ALL VERY IMPORTANT
0 1 2 3 4 5 6 7 8 9 l 0

Q that the movie is directed by a director you like?

0 l 2 3 4 5 6 7 8 9 10
Q the story type of the movie? (that is, Drama. Western. Action Adventure, Comedy. Sci-F1 etc.)

0 l 2 3 4 5 6 7 8 9 10
Q the ratings of the movie? (that is. G; PG; PG-l3; R; NC-l7 or X)

0 l 2 3 4 5 6 7 8 9 10
Q the reviews on radio or television. or newspapers. or magazines?

0 l 2 3 4 5 6 7 8 9 10
Q the previews you see at the movies or on video?

0 l 2 3 4 5 6 7 8 9 [O
Q that you know about the story from a novel, play. or a television show?

0 1 2 3 4 S 6 7 8 9 10
Q the recommendations of other people?

0 l 2 3 4 5 6 7 8 9 10
Q the nomination or winning of awards like Oscar, or Golden Globe?

0 1 2 3 4 5 6 7 8 9 10
Q the advertising in the media? (that is. TV and radio commercials. newspaper ads etc..)

0 1 2 3 4 5 6 7 8 9 10

Q hearing about the movie from the stars who appear on programs like the Today Show, or late
night TV shows?

0 l 2 3 4 5 6 7 8 9 10

Q that the current movie is a sequel of a one you have seen and liked?

0 l 2 3 4 5 6 7 8 9 10

10.

I42

. Some people say that it is easier to decide to rent a movie than to go a theater to see it. Do you,

Strongly Agree, A gree, Neither agree nor disagree, Disagree or Strongly Disagree?

IDSA 2OA 3DN 4CD 5OSD 9QDK/RF

. Suppose a movie was available on video at the same time it was released to the theaters, how

often would you rent the video instead of going to the theater? Would you say it is...

1 C] Very Often 2 C] Often 3 [3 Sometimes 4 C] Rarely 5 [3 Never 9 D DK/RF

. Would you be willing to pay more for such an early release than what you usually pay now to

rent a video?

IDYES 2QNO 9DDK/RF

. If you were to single out the most important reason for going to a theater to see a movie,

 

what would it be? [WRITE IN VERBATIM]

What do you like the least about going to a theater to see a movie? [WRITE IN VERBATIM]

NOW A FEW QUESTIONS ABOUT CABLE TV AND VIDEO RENTAL

11.

12.

l3.

14.

18 cable TV available where you live? Q 1 [3 YES
2 C) NO QQQ SKIPTO Q #18 9 C] DK/RF

Doyou subscribetocable TV? Q 1C] YES
2C] NoQQQ SKIPTOQ#l8 9D DK/RF

Do you subscribe to any of the following movie channels? [check all that apply]

[3 HBO D Cinemax D Showtime D Disney O TMC (The Movie Channel)
(3 DK/RF C] NONE Q SKIPTOQIIS

In an average month, about how many movies do you watch on these channels?

0 l 2 3 4 5 6 7 8 9 10+

15.

l6.

17.

18.

143

Is pay-per-View (request TV) available where you live?

IDYES 2DN0QQQ SKIPTOQ#18 9DDK/RF

Have you (or anyone in your household) ever requested, and paid for, any movies on pay-
per-view?

ICJYES 2CJNOQQQ SKIPTOQ#18 9C]DK/RF

How often do you request movies on pay-per-View? Would you say it is about...

ID more than once a week
2 [3 once a week,

3 [3 once every two weeks.

3 [3 once a month, or

4 [:3 less than once a month?
9 [3 DK/RF

DO you have a... VCR? QQQQQ 1L3 Yes 2 C) No 9 C] DK/RF

How about a Laser Disc player? =9 1 D Yes 2 [:J No 9 C] DK/RF

[IF 'NO’ TOTHE BOTH ASKTHE RESPONDENT RENT MOVIE VIDEOS. IF ’YES’ GOTO THE NEXT; IF ’NO' SKIPTO #31]

19.

20.

21.

22.

In an average month about how many movies do you rent? [Videos or laser discs]

0 1 2 3 4 5 6 7 8 9 10+

 

When you or someone in your household rents a movie,
about how many people watch it? QQQ

 

 

 

Can you recall the last movie you rented? 2 C] NO :9 SKIPTOQ #2304130 PAGE]
1C] YES 4' [WRITETHE NAME BELOW] 9 C3 DK/RF

What attracted you to that movie?

 

23.

24

144

Following questions deal with the way people rent movie videos. For each question the
answer can range fiom Always to Never. To begin with...

How often do you go to the Video store with a particular movie in mind to rent? Would you
say it is...

C] Always [3 Almost always 1:] About half the time D Rarely 0 Never {3 DK/RF

How often do you rent something else if the movie you are looking for is not available? Would
you say it is...

D Always C] Almost always D About half the time D Rarely [3 Never C] DK/RF

How often do you go to another video store if the movie you are looking for is not available?
Would you say it is...

C) Always C] Almost always [:1 About half the time C] Rarely 1:) Never D DK/RF

How often do you rent “newly released” movies? (eg. the tOp 30 titles) Would you say it is...

[3 Always (3 Almost always C3 About half the time D Rarely O Never 0 DK/RF

How often do you rent the movies that are not in the ‘new releases’ shelves? Would you say it
is...

[3 Always [:3 Almost always C] About half the time [:1 Rarely C] Never [3 DK/RF

How often do you rent the movies you have already seen in the theater? Would you say
it is...

0 Always C] Almost always D About half the time D Rarely C] Never [3 DK/RF
About how often do you wait to see a movie on video rather than going to a theater? Would
you say it is...

D Always [:3 Almost always C] About half the time D Rarely [3 Never D DK/RF

Have you purchased any movie videos in the past? Q 1 C] Yes 2 C] No 9 D DK/RF

27.

28.

30.

145

With whom do you usually watch movies at home? (on video)

1 [:1 With friends

2 [3 With acquaintance
3 (3 With family

4 [3 With a date

5 [3 Alone

6 C) Other

9O DK/RF

 

Thinking about all the movies you have seen over the past six months in the

theaters and on video, about what percentage did you see at the theaters? Q
[With 100% being the total]

 

 

 

If you were to single out the most important reason for renting a movie rather than going to
the theater to see it, what would it be?

 

What do you like the least about renting a movie?

 

 

 

 

Including the concession prices. about how much do you spend when you

go out to see a movie in a theater? Q
[may also including baby sitting costs and the like]

 

 

 

Do you think you would go to the movies (in theaters) more often than you do now if the
ticket prices were lower?

ID Yes 2D No 9C3 DK/RF

F inally, a few more questions about yourself to help us interpret the survey data.

31.

Do you currently work?

ID Yes QQQQQQ D FullTime OR B PartTime
20 No

9C3 DK/RF

32.

33.

34.

35.

36.

37.

146

Are you... 1C] Single 2C] Married, or 3D Living together? 90 DK/RF

Do you have children under 16 living with you now?

 

10 Yes c$QQHowmany?Q
2D No
9C] DK/RF

 

 

 

Including yourself , how many people live in your household? Q

Your age is between...[READ OPTIONS]

IC] 1620
2(3 2124
3C] 2529
4C) 3039
5C] 4049
6D 5059
7C] 600rolder
9C) DK/RF

Your education level is...[READ OPTIONS]

1 [3 Less than high school

2 C] High School completed
3 [:3 Some College

4 D College graduate

5 [3 Post graduate

9C] DK/RF

 

 

 

 

Which of the following income categories best describe your household income?

1 [3 Under $15,000 per year
2 [:1 515.001 to $25,000

3 [3 $25,001 to $35,000

4 C) $35,001 to $45,000

5 C] $45,001 to 355.000

6 (:1 $55,001 to $65,000

7 C] $65,001 and above

9 C) DK/RF

Myalbfiwqmdorwwemand/Wyowm mowhfnryour tune

 

INTERVIEWER: Mark the gender of the respondent QQQ 1 [3 Male

2 C] Female

Appendix 2

147

 

 

 

TABLE 4.8
FACTOR MATRIX

VARIABLE FACTOR 1 FACTOR 2 FACTOR 3 FACTOR 4
SATRS .46913 .02971 .46503 .22735
DIRECTOR .44974 .41950 .26150 .39131
ST ORY'I'YP .09826 .38240 .41427 .14366
RATINGS .34314 .18298 .54158 .28238
REVIEWS .44470 .31951 .32191 .26148
PREVIEWS .12109 .69584 .29826 .20324
PRESTORY .48978 .45328 .19746 .16169
WORDOFM .42151 .24053 .241 73 .47668
AWARDS .72221 .18787 .01256 .16331
ADVTSNG .407 25 .57258 .33516 .19123
TVPROMOS .67342 .053“) .14012 .24665
SEQUEL .51631 .31872 .04912 .55632

 

148

 

 

 

 

TABLE 4.11
TOLERANCE AND VARIANCE INFLATION FACTORS
VARIABLES IN THE EQUATION VARIABLES NOT IN
VARIABLE TOLERANCE VIF VARIABLE TOLERANCE VIF

AGE .602263 1.660 SATRS .855528 1.169

DIRECTOR .991837 1 .008 ST ORY'TYP .930377 1 .075

ADV'TSNG .941977 1.062 RATINGS .90le 1.109

REVIEWS .876237 1.141 PREVIEWS .815575 1.226

INCOME .475057 2.105 PRESTORY 86% 1.151

MARITAL .531190 1.883 WORDOFM .879956 1.136
AWARDS .644385 1.552
TVPROMOS .877947 1.139
SEQUEL .956757 1.045
MOVICOST .768705 1.301
ADMSN$ .861889 1.160
CHILDREN .756679 1.322
HHHEADS .831450 1.203
EDULEVEL .821773 1.217
GENDER .928711 1.077

 

SMALLER TOLERANCE INDICATE A LINEAR COMBINATION WITH ANOTHER VARIABLE

Table 4.11 Cont’d...
COLLINIEARTTY DIAGNOSTICS

 

VARIANCE PROPORTIONS

 

Number Eigenval C Index Constant Director Reviews Advtsng Marital Age Income

 

1 5.71610 1.000 111158 .(XI651 .(XJ315 .00273 .00579 .00388 .(Xl379
2 .63862 2.992 .00460 .06902 .00682 .01386 .27866 .00517 .02361
3 .24194 4.861 .00645 .84719 .03751 .04026 .08725 .01470 111269
4 .16331 5.916 111422 .00970 .00025 .16217 .40415 .39646 .06760
5 .10974 7.217 .(XXIOI .00120 .10756 .02373 .21861 .2189] .84023
6 .08882 8.022 .0168] .00084 .78276 .20439 .00002 .26312 .05813
7 .04146 11.741 .%633 .06554 .06197 .55287 .01051 .09775 .00395

 

References

REFERENCES
Acheson. K. Maule. C. and Filleul E. “Folly of Quotas on Films and Television
Programmes." World Economy, December, 1989. pp. 515-524.
Allen, Robert C. (1985). “The Movies in Vaudeville: Historical Context of the
Movies as Popular entertainment.” in The American Film Industry. Balio.
Tino. ed. Madison, Wisconsin: The University of Wisconsin Press.

Austin, B. A.( 1981) "Rating the Movies." Journal of Popular Film and Television
7: 4. 384-399.

Austin. B. A.( 1981b) "Film Attendance: Why College Students Choose to See
Their Most Recent Film." Journal of popular Film and Television 9: 43-49.

Austin, B. A. (1982) "A Factor Analytic Study of Attitudes Toward Motion
Pictures." Journal of Social Psychology 117: 211-217.

Austin, B. A. (1984) "But Why This Movie?" Boxojfice February, 16-18.

Austin. B. A. (1984) "Motivations for Movie Attendance." Boxofl'ice October. 13-
16

Austin. Bruce A. (1989). Immediate Seating: A Look at Movie Audiences.
Belmont, California: Wadsworth.

Babbie, Earl. (1990) Survey Research Methods. Wadsworth California.

Balio, Tino. ed. (1985). The American Film Industry. Madison. Wisconsin: The
University of Wisconsin Press.

Bernstein. Carl (1990). “The Leisure Empire" Time. December 24. pp. 56-59.
Blalock. Hubert M. (1979) Social Statistics. McGraw-I-Iill, New York.

Blumler. J .G. (1979) "The Role of Theory in Uses and Gratifications studies."
Comrrumicatlon Research 6: 1. 9-36.

Blumler. J .G. and Katz. E. eds. (1974) The Uses of Mass Communications Sage
CA.

Bowman, G.W. and Farley. J.U. (1972) "TV Viewing: Application of a Formal
Choice Model." Applied Economics 4. 245-259.

Collins, J. et a1. "Predicting Cable Subscribership: Local Factors." Journal of
Broadcasting 27:2, 177-183.

Conant. Michael. (1960) Antitrust in the Motion Picture Industry: Economic and
Legal Analysis. Los Angeles: University of California Press.

149

150

Cooper-Martin. E. (1991) "Consumers and Movies: Some Findings on
Experimental Products." Advances in Consumer Research Vol. 18. pp.
372-378.

Corlis. R. (1992) "Binge and Purge at the 8.0." Time January 20. p.59
Corliss. Richard. (1990). Box-Office Brawn. Time. December 24. pp. 52-55.

Dillman. Don A. (1978) Mail and Telephone Surveys: The Total Design Method.
John Wiley 8: Sons. New York.

Dominick. Joseph R. (1987) “Film Economics and Film Content: 1964-1983" in
Bruce A. Austin (ed.). Current Research in Film: Audiences. Economics
and Law. Vol. 3. Norwood. NJ: Ablex. . ch. 9.

Donohew. L. et a1. (1987) "Social and Psychological Origins of Media Use: A
Lifestyle Analysis." Journal of Broadcasting and Electronic Media 31: 3.
255-278.

Ducey. R.V. "Predicting Market Segments in the Cable Industry: The Basic and
Pay Subscribers." Journal of Broadcasting 27:2. 155-161.

The Economist. (1993). “Multimedia: The Tangled Webs They Weave” October
16. pp 21-24.

The Economist. (1989) "The Entertainment Industry: Rasing the Stakes."
December. 1989.

Ehrenberg. A.S.C. (I968) "The Factor Analytic Search for Program Types."
Journal of Advertising Research 8:1. 55-63.

Elliott, R. W. and Quattlebaum. C. P. (1979) "Similarities in Patterns of Media
Use: A Cluster Analysis of Media Gratifications." The Western Journal of
speech communication 43: Winter 61-72.

Elliott. Richard. and Hamilton. Eleanor. (1991) "Consumer Choice Tactics and
Leisure Activities." International Journal of Advertising, 10. 325-332.

Faber. Ronald J. and Thomas C. O'Guinn (1984) "Effects of Media Advertising
and Other Sources on Movie Selection." Journalism Quarterly. 61
(Summer). pp. 371-377.

Fabrikant. Geraldine (1990). Hollywood Takes More Cues From Overseas. The
New York Times. May 25. p. A1.

Galloway, J .J . and Meek. F.L. (1981) "Audience Uses and Gratifications: An
Expectancy Model." Communication Research 8: 4. 435-449.

Glock. Charles (ed) (1967) Survey Research in the Social Sciences. Sage. New
York.

Goldwyn. Samuel. (1950) “Hollywood in the Television Age." Hollywood
Quarterly. vol. 4. p.145.

151

Gomery. Douglas (1991). Movie History: A Survey. Belmont. California:
Wadsworth.

Gomery. Douglas (1993). “The Contemporary American Movie Business.” in
Media Economics Theory and Practice. Alexander Alison. Owers James.
and Carveth Rod. ed. Hillsdale. New Jersey: Lawrence Erlbaum

Associates.

Guback. Thomas (1987). The “Evolution of the Motion Picture Theater Business
in the 1980s.” Journal of Communication Spring.

Guback. Thomas (1969). The International Film Industry. Bloomington: Indiana
University Press.

Gutman. J. (1978) "Television Viewer Types: A Q Analysis." Journal of
Broadcasting 23:4. 505-515.

Haley. J. (1951-52) "The appeal of the Motion Pictures." The Quarterly of Film
Radio and TV 6:361-374.

Hendricks. Gordon (1985). “The History of Kinetoscope.” in The American Film
Industry. Balio. Tino. ed. Madison. Wisconsin: The University of
Wisconsin Press. p. 53.

Hirschman. E.C. and Holbrook. MB. (1982) "Hedonic Consumption: Emerging
Concepts. Methods and Propositions." Journal of Marketing. 46
(Summer). pp. 92-101.

Holbrook. MB. and Hirschman. E. C. (1982) "The Experimental Aspects of
Consumption: Consumer Fantasies. Feelings and Fun." Journal of
Consumer Research 9: September. 132-140.

Howard. J .A. and Sheth. J.N. (1969) The Theory of Buyer Behavior John Wiley.
New York.

Howard. John A. Consumer Behavior: Application of Theory McGraw-Hill. New
York 1977

Huettig. Mae D. (1985). “Economic Control of the Motion Picture industry." in
The American Film Industry. Balio. Tino. ed. Madison. Wisconsin: The
University of Wisconsin Press.

Johnson. Michael D. (1984). "Consumer Choice Strategies for Comparing
Noncomparable Alternatives." Journal of Consumer Research. 11
(December). pp. 741-753.

Jowett. Garth (1976) Film.- The Democratic Art. Boston: Little. Brown and
Company.

Kaplan. A. (1963) The Conduct of Inquiry: Methodology for Behavioral Science.
Harper & Row. New York.

Katz. E. Blumler J .G. and Gurevitch. M. “Uses and Gratifications Research”
Public Opinion Quarterly. 37:509-537. 1973

152

Katz. E. (1979) "The Uses of Becker. Blumler. and Swanson." Communication
Research 6:1. 74-83.

LaRose. R. and Atkin. D. (1991) "Attributes of Movie Distribution Channels and
Consumer Choice." Journal of Media Economics Spring. 3-17.

Lazarsfeld. Paul. (1947 ) “Audience Research in the Movie Field.” Annals of the
American Academy of Political and social Science. vol. 254. pp. 160-168.

Lehmann. D.R. (1971) "Television Show Preference: Application of A Choice
model." Journal of Marketing Research 8: February. 47-55.

Leuvaen. J .V. (1981) "Expectancy Theory in Media and message Selection."
Communication Research 8:4. 425-434.

Liechtenstein. a. and Rosenfeld. L. (1983) "Uses and Misuses of Gratifications
Research: An Explication of Media Functions." Communication Research
10:1. 97-109.

Litman. Barry. (1979) “Predicting TV Ratings for Theatrical Movies." Journalism
Quarterly 53: Autumn. pp. 590-594: 694

Litman. Barry. (1982) “Decision Making in film Industry: The influence of the
TV Market.” Journal of Communication 32: Summer. 32-52

Litman. B.R. (1983) "Predicting Success of Theatrical Movies: An Empirical
Study." Journal of Popular Culture 16: Spring. 159-75.

Litman. B.R. and Kohl. L.(l989) "Predicting Financial Success of Motion
Pictures: The '808 Experience." Journal of Media Economics Fall. 1989
35-50.

Londoner. David J. (1985). “The Changing Economics of entertainment." in The
American Film Industry. Balio. Tino. ed. Madison. Wisconsin: The
University of Wisconsin Press.

Lowery. Tina M. “The Use of Diffusion Theory in Marketing: A Qualitative
Approach to Innovative Consumer Behavior" Advances in Consumer
Research (Vol 18) 1991 pp 644-649.

McDonald. D.G. (1983) "Investigating Assumptions of Media Dependency
Research." Communication Research 10:4, 509-528.

Merritt. Russell (1985). “Nickelodeon Theaters. 1905-1914: Building an
Audience for the Movies.” in The American Film Industry. Balio. Tino.
ed. Madison. Wisconsin: The University of Wisconsin Press.

Musun. C. (1969) The Marketing of Motion Pictures: Both sides of The Coin Chris
Musun Co., Los Angeles.

Nichols. P.M. (1992a) "If You Want to Watch 'J.F.K.' in Private. You'll Have a
Shorter Period to Wait Than Expected." The New York Times. April 23. p.
B2.

153

. ( 1992b) "Studios Pursue New Stratagies As They Try to Turn Their Best
Renters Into Best Sellers." The New York Times. May. 2. p. BS.

. (19920) "Viewers Want Hits. So Dealers Press for Speed But Offer
Fewer Titles." The New York Times. June. 18. p B4.

Nunnally, Jum C. (1978) Psychometric Theory. McGraw-Hill. New York.

Oppenheirn. A. N. (1966) Questionnaire Design and Attitude Measurement.
Basic Books. New York.

Olshavsky. R. W. and Granbois. D.H. (1979) "Consumer Decision Making-Fact
or Fiction?" Journal of Consumer Research. 6 (September). 93- 100.

Palmgreen. P. Peter L.C. James. G.H. and David M.H. “The Motivational
Framework of Moviegoing: Uses and Avoidances of Theatrical Films.” in
B.A. Austin. ed. Current Research in nlmAudiences. Economics and Law.
Noewood. N.J. Ablex Publishing. 1987, pp. 40-57.

Palmgreen. P. et al. (1981) “Gratification Discrepancies and News Program
Choice." Communication Research 8: 4. 451-478.

Palmgreen. P. and Rayburn. J .D.( 1979) "Uses and Gratifications and Exposure
to Public Television: A Discrepancy Approach." Communication Research
6:2. 155-180.

Platt. Richard. (1992) Film. Alfred A. Knoff: New York.

Pucely. J .M. et al. (1988) "A Comparison of Involvement Measures for the
Purchase and Consumption of Pre-Recorded Music." Advances in
Consumer Research Volume 15. pp. 37-42.

Quigley Publishing Company. Inc.. (1991) International Motion Picture Alumnae.
New York. N.Y.

Quigley Publishing Company. Inc.. (1992) International Television & Video
Almanac. New York. N.Y.

Rubin. A.M. (1983) "Television Uses and Gratifications: The Interactions of
Viewing Patterns and Motivations." Journal of Broadcasting 27:1 Winter.
37 -51.

Rogers. Everett M. (1983) Diffusion of Innovations. 3rd ed. New York: Free
Press.

Rubin. A.M. (1981) "An Examination of Television Viewing Motivations."
Communication Research 8:2. 141-165.

Rubin. A.M. and Windahl. S. (1986) "The Uses and Dependency Model of Mass
Communication." Critical studies in Mass Communication 3: 184-99.

Sheth. J .N. (1979) "The Surpluses and Shortages in Consumer Behavior Theory
and Research." Journal of the Academy of Marketing Science. 7(4). 414-
427.

154

Simonet. Thomas. “Conglomerates and Content: Remakes. Sequels. and Series
in the New Hollywood.” in Bruce A. Austin (ed.). Current Research in
Film: Audiences. Economics and Law. Vol. 3. Norwood. NJ: Ablex. 1987.
ch. 10.

Sochay. Scott. and Litman. Barry. “Export Markets and the U.S. Motion Picture
Industry.” Journal of Media economics. Fall 1992.

Steiner. PD. (1952) "Program Patterns and Preferences. and the Workability of
Competition in radio Broadcasting." The Quarterly journal of Economics
Vol LXVI. 194-223.

Stuart. Fedrick (1975) The Efi‘ects of Television on the Motion Picture and Radio
Industries New York: Arno Press

Swanson. D.L. "Political Communication Research and Uses and Gratifications
Model: A Critique." Comrruinlcation Research 6:1. 37-53.

Wakshlag. J .J . et al. "Television News Viewing and Network Affiliation
Changes." Journal of Broadcasting 27:1 Winter 53-67.

Webster. J .G. and Wakshlag. J .J . "A Theory of Television Program Choice."
Communication Research 10:4, 430-446.

Wells. W. ( 1969) "The Rise and Fall of Television Program Types." Journal of
Advertising Research 9:3. 21-27.

U.S. Department of Commerce. (1992). U.S. Industrial Outlook 1992

Veronis. Suhler 8: Associates. Communications Industry Forecast. July 1992.
Veronis. Suhler 8: Associates Communication Industry Forecast. July 1990.
Vroom. V.H. (1964) Work and Motivation John Wiley. New York

Vogel. Harold L. (1986) Entertainment Industry Economics: A Guide for Financial
Analysis. Cambridge: Cambridge University Press.

Zufryden. F.S. (1976) "Patterns of TV Progam Selection." Journal of Advertising
Research 16:6. 43-47.

”iiiiiiliii‘iiilr